TVCG2026

Abstract:
Normal reconstruction is crucial in non-line-of-sight (NLOS) imaging, as it provides key geometric and lighting information about hidden objects, which significantly improves reconstruction accuracy and scene understanding. However, jointly estimating normals and albedo expands the problem from matrix-valued functions to tensor-valued functions that substantially increasing complexity and computational difficulty. In this paper, we propose a novel joint albedo-surface reconstruction method, which utilizes the shape operator to control the variation rate of the normal field. It is the first attempt to apply regularization methods to the reconstruction of surface normals for hidden objects. By improving the accuracy of the normal field, it enhances detail representation and achieves high-precision reconstruction of hidden object geometry. The proposed method demonstrates robustness and effectiveness on both synthetic and experimental datasets. On transient data captured within 15 seconds, our surface normal-regularized reconstruction model produces more accurate surfaces than recently proposed methods and is 30 times faster than the existing surface reconstruction approach.

Abstract:
Resampling of warped images has been a topic of research for a long time but only seldomly has focused on theoretically exact resampling. We present a resampling method for minification, applied on the texture mapping function of a 3D graphics pipeline, that is derived from sampling theory without making any approximations. Our method supports freely selectable 2D integratable prefilter (anti-aliasing) functions and uses a 2D box reconstruction filter. We have implemented our method both for CPU and GPU (OpenGL) using multiple prefilter functions defined by piece-wise polynomials. The correctness of our exact resampling method has been made plausible by comparing texture mapping results of our method with those of extreme supersampling. We additionally show how the prefilter of our method can also be applied for high quality polygon edge anti-aliasing. Since our proposed method does not use any approximations, up to numerical precision, it can be used as a reference for approximate texture mapping methods.

Abstract:
Voronoi treemaps are used to depict nodes and their hierarchical relationships simultaneously. However, in addition to the hierarchical structure, data attributes, such as co-occurring features or similarities, frequently exist. Examples include geographical attributes like shared borders between countries or contextualized semantic information such as embedding vectors derived from large language models. In this work, we introduce a Voronoi treemap algorithm that leverages data similarity to generate neighborhood-preserving treemaps. First, we extend the treemap layout pipeline to consider similarity during data preprocessing. We then use a Kuhn-Munkres matching of similarities to centroidal Voronoi tessellation (CVT) cells to create initial Voronoi diagrams with equal cell sizes for each level. Greedy swapping is used to improve the neighborhoods of cells to match the data's similarity further. During optimization, cell areas are iteratively adjusted to their respective sizes while preserving the existing neighborhoods. We demonstrate the practicality of our approach through multiple real-world examples drawn from infographics and linguistics. To quantitatively assess the resulting treemaps, we employ treemap metrics and measure neighborhood preservation.

Abstract:
Principal Component Analysis (PCA) is perhaps the most popular linear projection technique for dimensionality reduction. We consider PCA under the assumption that the high-dimensional data points are equipped with Gaussian uncertainty. Several approaches to such uncertainty-aware PCA have been developed recently in the visualization community. Since PCA is a discontinuous map, a small uncertainty in the data points can result in a huge uncertainty in the projected points. We show that the uncertainty of the data points also creates uncertainty in the eigenvectors of the covariance matrix that defines the PCA projection. We present a closed-form expression to quantify eigenvector uncertainty. Based on this, we propose a 3D glyph that supports the decision whether existing solutions for uncertainty-aware PCA are sufficient, or whether a more expensive sampling-based approach is required. We apply our approach to several test data sets.

Abstract:
Selecting the appropriate dimensionality reduction (DR) technique and determining its optimal hyperparameter settings that maximize the accuracy of the output projections typically involves extensive trial and error, often resulting in unnecessary computational overhead. To address this challenge, we propose a dataset-adaptive approach to DR optimization guided by structural complexity metrics. These metrics quantify the intrinsic complexity of a dataset, predicting whether higher-dimensional spaces are necessary to represent it accurately. Since complex datasets are often inaccurately represented in two-dimensional projections, leveraging these metrics enables us to predict the maximum achievable accuracy of DR techniques for a given dataset, eliminating redundant trials in optimizing DR. We introduce the design and theoretical foundations of these structural complexity metrics. We quantitatively verify that our metrics effectively approximate the ground truth complexity of datasets and confirm their suitability for guiding dataset-adaptive DR workflow. Finally, we empirically show that our dataset-adaptive workflow significantly enhances the efficiency of DR optimization without compromising accuracy.

Abstract:
Hex-dominant mesh generation has recently received increasing attention from researchers and the simulation community due to its robustness compared to pure hex-mesh generation techniques. In this work, we present a first structure-informed simplification framework, aiming to reduce the number of non-hex cells from the hex-dominant meshes. Our framework extracts individual sub-structures via parallel relations from the input hex-dominant mesh, decomposes self-tangent and self-intersecting sub-structures for structure complexity control, and collapses certain sets of edges that are adjacent to particular non-hex cells to remove them. We design a filtering and ranking strategy to select edges and sub-structures for collapsing. To better understand the complexity of sub-structure configurations, we introduced a novel relation graph that captures the connections between edges and between edges and sub-structures. Additionally, we designed a smoothing algorithm for hex-dominant meshes that enhances hex cell quality, even in meshes containing various cell types. We evaluate the effectiveness of our framework by applying it to many hex-dominant meshes produced by three state-of-the-art hex-dominant meshing techniques. Our results achieve various levels of reduction in the number of non-hex cells of the input meshes, affirming the applicability of our framework for improving the hex-dominant meshes.

Abstract:
Projector compensation seeks to correct geometric and photometric distortions that occur when images are projected onto nonplanar or textured surfaces. However, most existing methods are highly setup-dependent, requiring fine-tuning or retraining whenever the surface, lighting, or projector-camera pose changes. Progress has been limited by two key challenges: (1) the absence of large, diverse training datasets and (2) existing geometric correction models are typically constrained by specific spatial setups; without further retraining or fine-tuning, they often fail to generalize directly to novel geometric configurations. We introduce SIComp, the first Setup-Independent framework for full projector Compensation, capable of generalizing to unseen setups without fine-tuning or retraining. To enable this, we construct a large-scale real-world dataset spanning 277 distinct projector-camera setups. SIComp adopts a co-adaptive design that decouples geometry and photometry: A carefully tailored optical flow module performs online geometric correction, while a novel photometric network handles photometric compensation. To further enhance robustness under varying illumination, we integrate intensity-varying surface priors into the network design. Extensive experiments demonstrate that SIComp consistently produces high-quality compensation across diverse unseen setups, substantially outperforming existing methods in terms of generalization ability and establishing the first generalizable solution to projector compensation. The code and dataset are available on our project page: https://hai-bo-li.github.io/SIComp/.

Abstract:
The reconstruction of realistic and precise human meshes in world coordinates is facilitated by considering scene information. Challenges related to accuracy, robustness, and computation time are faced by existing absolute human mesh recovery methods. In this paper, a one-stage model for absolute human mesh recovery with superior reconstruction precision and inference speed is presented. The proposed one-stage model is composed of two parallel branches to achieve root position estimation and human mesh regression. To effectively connect the two branches, a scene-image information aggregation module is designed. The accuracy of the estimated human meshes is improved and the end-to-end training of the whole model is facilitated by this module. Experiments are conducted on three diverse datasets, and a GMPJPE decrease of 72.3 mm/27.32% and an MPJPE reduction of 25.6 mm/27.26% are achieved by the proposed method with the lowest inference time compared to previous SOTA methods.

Abstract:
We investigate tasks that can be accomplished with unlabeled graphs, which are graphs with nodes that do not have persistent or semantically meaningful labels attached. New visualization techniques to represent unlabeled graphs have been proposed, but more understanding of unlabeled graph tasks is required before these techniques can be adequately evaluated. Some network visualization tasks apply to both labeled and unlabeled graphs, but many do not translate between these contexts. We propose a data abstraction model that distinguishes the Unlabeled context from the increasingly semantically rich Labeled, Attributed, and Augmented contexts. We filter tasks collected and gleaned from the literature according to our data abstraction and analyze the surfaced tasks, leading to a taxonomy of abstract tasks for unlabeled graphs. Our task taxonomy is organized according to the Target data under consideration, the Action intended by the user, and the Scope of the data at play. We show the descriptive power of this task abstraction by connecting to concrete examples from previous frameworks, and connecting these abstractions to real-world problems. To showcase the evaluative power of the taxonomy, we perform a preliminary assessment across 6 different network visualization idioms for each task. For each combination of task and visual encoding, we consider the effort required from viewers, the likelihood of task success, and how both factors vary between small-scale and large-scale graphs.

Abstract:
Projector photometric compensation corrects color distortions introduced by surface texture, reflection, and ambient lighting. Existing deep learning-based methods usually require professional scene-specific data collection and lack consideration for perceptual quality. To address this limitation, we present a diffusion-based photometric compensation method that reconstructs compensation images under photometric and content-aware guidance. Specifically, we fi rst mo del th e ph otometric distortions introduced during projection as environment-dependent additive noise, thereby reformulating the photometric compensation problem as a denoising task with physical constraints. Next, we introduce a diffusion model, which generates compensation images by following an additive trajectory to iteratively remove the noise. Finally, to accurately estimate the noise at each timestep, by analyzing the factors that contribute to distortions in the physical process of projection and capturing, we design a noise estimation network that incorporates features of both photometry-aware and content conditions. Experiments show that our method achieves superior visual performance in unknown scenarios, thereby exhibiting significant practical advantages over prior art. Our source code is available at https://github.com/cyxwang/DiffPC.

Abstract:
This short paper presents a general approach for computing robust Wasserstein barycenters (Agueh et al. 2011), (Turner et al. 2014),(Vidal et al. 2020) of persistence diagrams. The classical method consists in computing assignment arithmetic means after finding the optimal transport plans between the barycenter and the persistence diagrams. However, this procedure only works for the transportation cost related to the qq-Wasserstein distance W_qWq when q=2q=2. We adapt an alternative fixed-point method (Tanguy et al. 2025) to compute a barycenter diagram for generic transportation costs (q > 1q>1), in particular those robust to outliers, q \in (1,2)q∈(1,2). We show the utility of our work in two applications: (i) the clustering of persistence diagrams on their metric space and (ii) the dictionary encoding of persistence diagrams (Sisouk et al. 2024). In both scenarios, we demonstrate the added robustness to outliers provided by our generalized framework.

Abstract:
Occlusion culling is a cornerstone of real-time rendering, particularly in mobile games where limited GPU bandwidth demands highly efficient scene management. At the heart of occlusion culling lies the use of simplified proxy geometry—called occluders—that approximate scene geometry for rapid visibility testing. However, producing high-quality occluders that are low in polygon count, conservative in coverage, and tightly aligned with the original geometry remains a manual and labor-intensive process. In this paper, we present a fast and fully automated two-stage approach for robust occluder generation tailored to real-world game assets. Our method begins with a novel strategy for inward offset mesh computation, followed by a conservative simplification step leveraging a new variant of Quadric Error Metrics (QEM). This approach effectively handles noisy and topologically complex inputs, generating production-ready occluders in seconds. Extensive experiments on a wide range of asset types demonstrate that our technique achieves aggressive triangle reduction while preserving critical occlusion fidelity. By offering a practical and scalable solution, our method bridges the gap between academic research and demanding needs for game development.

Abstract:
To integrate digital humans into everyday life, there is a strong demand for generating high-quality, fine-grained disentangled 3D avatars that support expressive animation and simulation capabilities, ideally from low-cost textual inputs. Although text-driven 3D avatar generation has made significant progress by leveraging 2D generative priors, existing methods still struggle to fulfill all these requirements simultaneously. To address this challenge, we propose DreamBarbie, a novel text-driven framework for generating animatable 3D avatars with separable shoes, accessories, and simulation-ready garments, truly capturing the iconic “Barbie doll” aesthetic. The core of our framework lies in an expressive 3D representation combined with appropriate modeling constraints. Unlike prior methods, we use G-Shell to uniformly model watertight components (e.g., bodies, shoes) and non-watertight garments. By reformulating boundaries as euclidean field intersections instead of manifold geodesics, we propose an SDF-based initialization and a hole regularization loss that together achieve a 100×100× speedup and stable open topology without image input. These disentangled 3D representations are then optimized by specialized expert diffusion models tailored to each domain, ensuring high-fidelity outputs. To mitigate geometric artifacts and texture conflicts when combining different expert models, we further propose several effective geometric losses and strategies. Extensive experiments demonstrate that DreamBarbie outperforms existing methods in both dressed human and outfit generation. Our framework further enables diverse applications, including apparel combination, editing, expressive animation, and physical simulation.

Abstract:
We present RopStitch, an unsupervised deep image stitching framework with both robustness and naturalness. To ensure the robustness of RopStitch, we propose to incorporate the universal prior of content perception into the image stitching model by a dual-branch architecture. It separately captures coarse and fine features and integrates them to achieve highly generalizable performance across diverse unseen real-world scenes. Concretely, the dual-branch model consists of a pretrained branch to capture semantically invariant representations and a learnable branch to extract fine-grained discriminative features, which are then merged into a whole by a controllable factor at the correlation level. Besides, considering that content alignment and structural preservation are often contradictory to each other, we propose a concept of virtual optimal planes to relieve this conflict. To this end, we model this problem as a process of estimating homography decomposition coefficients, and design an iterative coefficient predictor and minimal semantic distortion constraint to identify the optimal plane. This scheme is finally incorporated into RopStitch by warping both views onto the optimal plane bidirectionally. Extensive experiments across various datasets demonstrate that RopStitch significantly outperforms existing methods, particularly in scene robustness and content naturalness.

Abstract:
In the visual analysis of hierarchical data, a main challenge is comparing data attributes both within and between different levels of the hierarchy. Such tasks are typically addressed using aggregated data, where the attribute of a parent node is calculated from the attributes of its children. Our review of existing literature on visualization methods that encode the hierarchy implicitly (e.g., icicle plots, treemaps) shows that most approaches rely on additive aggregation. Proportional aggregation, where a parent is assigned a weighted average of the values of its children, remains unexplored although relevant in practice. We introduce stalactite plots, a visualization technique that displays proportional aggregation and supports visual value comparison. Our empirical evaluation (N = 148, N = 50) shows that, with some explanation, stalactite plots are as easily understood as established visualization techniques for hierarchical data. Furthermore, for large datasets, participants are faster and more accurate using our approach.

Abstract:
While recent advancements have demonstrated remarkable progress in general 3D shape generation, the challenge of automatically generating wearable 3D assets remains largely unexplored. To address this gap, we present BAG — a Body-aligned Asset Generation method that produces 3D wearable assets which can be automatically fitted onto given 3D human bodies. This is achieved by controlling the 3D generation process using human body shape and pose information. Specifically, we first construct a general single-image-to-consistent-multi-view diffusion model, and train it on the large-scale Objaverse dataset to ensure diversity and generalizability. We then train a body-conditioned multi-view ControlNet to guide the generator toward producing body-aligned multi-view images. The control signal leverages multi-view 2D projections of the target human body, where pixel values represent the XYZ coordinates of the body surface in a canonical space. The resulting body-conditioned multi-view diffusion outputs body-aligned images, which are subsequently fed into a native 3D diffusion model to reconstruct the 3D shape of the asset. Finally, we recover the similarity transformation using multi-view silhouette supervision and mitigate asset-body penetration using physics-based simulation, ensuring accurate asset fitting onto the target body. Experimental results demonstrate that our method significantly outperforms existing approaches in terms of prompt adherence, shape diversity, and shape quality.

Abstract:
Modeling 3D language fields with Gaussian Splatting for open-ended language queries has recently garnered increasing attention. However, recent 3DGS-based models leverage view-dependent 2D foundation models to refine 3D semantics but lack a unified 3D representation, leading to view inconsistencies. Additionally, inherent open-vocabulary challenges cause inconsistencies in object and relational descriptions, impeding hierarchical semantic understanding. In this paper, we propose Hi-LSplat, a view-consistent Hierarchical Language Gaussian Splatting work for 3D open-vocabulary querying. To achieve view-consistent 3D hierarchical semantics, we first lift 2D features to 3D features by constructing a 3D hierarchical semantic tree with layered instance clustering, which addresses the view inconsistency issue caused by 2D semantic features. Besides, we introduce instance-wise and part-wise contrastive losses to capture all-sided hierarchical semantic representations. Notably, we construct two hierarchical semantic datasets to better assess the model’s ability to distinguish different semantic levels. Extensive experiments highlight our method’s superiority in 3D open-vocabulary segmentation and localization. Its strong performance on hierarchical semantic datasets underscores its ability to capture complex hierarchical semantics within 3D scenes.

Abstract:
This work aims at a challenging task: human action-reaction synthesis, i.e., generating human reactions conditioned on the action sequence of another person. Currently, autoregressive modeling approaches with vector quantization (VQ) have achieved remarkable performance in motion generation tasks. However, VQ has inherent disadvantages, including quantization information loss, low codebook utilization, etc. In addition, while dividing the body into separate units can be beneficial, the computational complexity needs to be considered. Also, the importance of mutual perception among units is often neglected. In this work, we propose MARRS, a novel framework designed to generate coordinated and fine-grained reaction motions using continuous representations. Initially, we present the Unit-distinguished Motion Variational AutoEncoder (UD-VAE), which segments the entire body into distinct body and hand units, encoding each independently. Subsequently, we propose Action-Conditioned Fusion (ACF), which involves randomly masking a subset of reactive tokens and extracting specific information about the body and hands from the active tokens. Furthermore, we introduce Mutual Unit Modulation (MUM) to facilitate interaction between body and hand units by using the information from one unit to adaptively modulate the other. Finally, for the diffusion model, we employ a compact MLP as a noise predictor for each distinct body unit and incorporate the diffusion loss to model the probability distribution of each token. Both quantitative and qualitative results demonstrate that our method achieves superior performance.

Abstract:
Traditional analytical reflectance models, while compact and interpretable, lack the capacity to accurately represent physical measurements. Recent neural models, which closely fit input data, are less generalizable and often more expensive to store and evaluate. To combine the strengths and overcome the limitations of these two classes of models, we present neural enhancement, a novel framework to boost an input analytical appearance model, by identifying and replacing its key computational nodes/operators with small-scale multi-layer perceptrons. This allows us to leverage the computational graph structure of the original model, while improving its expressiveness at a modest cost. To make the enhancement computationally tractable, we propose a hypercube-based search to automatically and efficiently identify the node(s) and/or operator(s) to be replaced towards maximal gain in a differentiable fashion. We enhance a number of common analytical BRDF models. The results are, at once accurate, compact and efficient, and compare favorably with state-of-the-art work on fitting measured reflectance. Finally, our models are fully compatible with standard rasterization or ray-tracing pipeline.

Abstract:
We propose a novel framework for Russian Roulette and Splitting (RRS) tailored to wavefront path tracing, a highly parallel rendering architecture that processes path states in batched, stage-wise execution for efficient GPU utilization. Traditional RRS methods, with unpredictable path counts, are fundamentally incompatible with wavefront’s preallocated memory and scheduling requirements. To resolve this, we introduce a normalized RRS formulation with a bounded path count, enabling stable and memory-efficient execution. Furthermore, we pioneer the use of neural networks to learn RRS factors, presenting two models: NRRS and AID-NRRS. At a high level, both feature a carefully designed RRSNet that explicitly incorporates RRS normalization, with only subtle differences in their implementation. To balance computational cost and inference accuracy, we introduce Mix-Depth, a path-depth-aware mechanism that adaptively regulates neural evaluation, further improving efficiency. Extensive experiments demonstrate that our method outperforms traditional heuristics and recent RRS techniques in both rendering quality and performance across a variety of complex scenes.

Abstract:
With the revolution of generative AI, video-related tasks have been widely studied. However, current state-of-the-art video models still lag behind image models in visual quality and user control over generated content. In this paper, we introduce TokenWarping, a novel framework for temporally coherent video translation. Existing diffusion-based video editing approaches rely solely on key and value patches in self-attention to ensure temporal consistency, often sacrificing the preservation of local and structural regions. Critically, these methods overlook the significance of the query patches in achieving accurate feature aggregation and temporal coherence. In contrast, TokenWarping leverages complementary token priors by constructing temporal correlations across different frames. Our method begins by extracting optical flows from source videos. During the denoising process of the diffusion model, these optical flows are used to warp the previous frame’s query, key, and value patches, aligning them with the current frame’s patches. By directly warping the query patches, we enhance feature aggregation in self-attention, while warping the key and value patches ensures temporal consistency across frames. This token warping imposes explicit constraints on the self-attention layer outputs, effectively ensuring temporally coherent translation. Our framework does not require any additional training or fine-tuning and can be seamlessly integrated with existing text-to-image editing methods. We conduct extensive experiments on various video translation tasks, demonstrating that TokenWarping surpasses state-of-the-art methods both qualitatively and quantitatively. Video demonstrations are available in supplementary materials.

Abstract:
Diffusion-based generative models have achieved remarkable success in image restoration by learning to iteratively refine noisy data toward clean signals. Inspired by this progress, recent efforts have begun exploring their potential in 3D domains. However, applying diffusion models to point cloud denoising introduces several challenges. Unlike images, clean and noisy point clouds are characterized by structured displacements. As a result, it is unsuitable to establish a transform mapping in the forward phase by diffusing Gaussian noise, as this approach disregards the inherent geometric relationship between the point sets. Furthermore, the stochastic nature of Gaussian noise introduces additional complexity, complicating geometric reasoning and hindering surface recovery during the reverse denoising process. In this paper, we introduce a deterministic noise-free diffusion framework that formulates point cloud denoising as a two-phase residual diffusion process. In the forward phase, directional residuals are injected into clean surfaces to construct a degradation trajectory that encodes both local displacements and their global evolution. In the reverse phase, a U-Net-based network iteratively estimates and removes these residuals, effectively retracing the degradation path backward to recover the underlying surface. By decomposing the denoising task into directional residual computation and sequential refinement, our method enables faithful surface recovery while mitigating common artifacts such as over-smoothing and under-smoothing. Extensive experiments on synthetic and real-world datasets demonstrate that our method achieves state-of-the-art performance in both quantitative metrics and visual quality.

Abstract:
Point cloud normal estimation is a fundamental task in 3D geometry processing, playing a crucial role in applications such as 3D reconstruction, object recognition, and surface analysis. While recent learning-based methods achieve notable advancements in normal prediction, they often overlook the critical aspect of equivariance. This oversight leads to inefficient learning of symmetric patterns inherent in geometric data. To address this issue, we propose E^33-Net, an innovative neural network architecture designed to inherently achieve equivariance for normal estimation. We introduce an efficient random frame method, which significantly reduces the training resources required for this task to just 1/8 of previous work, while simultaneously enhancing prediction accuracy. Furthermore, we design a Gaussian-weighted loss function and a receptive-aware inference strategy that effectively leverage the local properties of point clouds, ensuring more precise and reliable normal estimation. Our method demonstrates superior performance across both synthetic and real-world datasets, consistently outperforming current state-of-the-art techniques by a substantial margin. Specifically, we achieve a 4% improvement in RMSE on the PCPNet dataset, 2.67% on the SceneNN dataset, and 2.44% on the FamousShape dataset, highlighting the robustness and scalability of E^33-Net in diverse environments.

Abstract:
We investigate the perceived visual complexity (VC) in data visualizations using objective image-based metrics. We collected VC scores through a large-scale crowdsourcing experiment involving 349 participants and 1,800 visualization images. We then examined how these scores align with 12 image-based metrics spanning pixel-based and statistic-information-theoretic (clutter), color, shape, and our two new object-based metrics (meaningful-color-count (MeC) and text-to-ink ratio (TiR)). Our results show that both low-level edges and high-level elements affect perceived VC in visualization images; the number of corners and distinct colors are robust metrics across visualizations. Second, feature congestion, a statistical information-theoretic metric capturing color and texture patterns, is the strongest predictor of perceived complexity in visualizations rich in the same continuous color/texture stimuli; edge density effectively explains VC in node-link diagrams. Additionally, we observe a bell-curve effect for texts: increasing TiR initially reduces complexity, reaching an optimal point, beyond which further text increases VC. Our quantification model is also interpretable-enabling metric-based explanations-grounded in the VisComplexity2K dataset, bridging computational metrics with human perceptual responses. The preregistration is available at osf.io/5xe8a. osf.io/bdet6 has the dataset and analysis code.

Abstract:
Visualization grammars from ggplot2 to Vega-Lite are based on the Grammar of Graphics (GoG), our most comprehensive formal theory of visualization. The GoG helped expand the expressive gamut of visualization by moving beyond fixed chart types and towards a design space of composable operators. Yet, the resultant design space has surprising limitations, inconsistencies, and cliffs - even seemingly simple charts like mosaics, waffles, and ribbons fall out of scope of most GoG implementations. To author such charts, visualization designers must either rely on overburdened grammar developers to implement purpose-built mark types (thus reintroducing the issues of typologies) or drop to lower-level frameworks. In response, we present GoFish: a declarative visualization grammar that formalizes Gestalt principles (e.g., uniform spacing, containment, and connection) that have heretofore been complected in GoG constructs. These graphical operators achieve greater expressive power than their predecessors by enabling recursive composition: they can be nested and overlapped arbitrarily. Through a diverse example gallery, we demonstrate how graphical operators free users to arrange shapes in many different ways while retaining the benefits of high-level grammars like scale resolution and coordinate transform management. Recursive composition naturally yields an infinite design space that blurs the boundary between an expressive, low-level grammar and a concise, high-level one. In doing so, we point towards an updated theory of visualization, one that is open to an innumerable space of graphic representations instead of limited to a fixed set of “good” designs.

Abstract:
Graph querying is the process of retrieving information from graph data using specialized languages (e.g., Cypher), often requiring programming expertise. Visual Graph Querying (VGQ) streamlines this process by enabling users to construct and execute queries via an interactive interface without resorting to complex coding. However, current VGQ tools only allow users to construct simple and specific query graphs, limiting users' ability to interactively express their query intent, especially for underspecified query intent. To address these limitations, we propose Envisage, an interactive visual graph querying system to enhance the expressiveness of VGQ in complex query scenarios by supporting intuitive graph structure construction and flexible parameterized rule specification. Specifically, Envisage comprises four stages: Query Expression allows users to interactively construct graph queries through intuitive operations; Query Verification enables the validation of constructed queries via rule verification and query instantiation; Progressive Query Execution can progressively execute queries to ensure meaningful querying results; and Result Analysis facilitates result exploration and interpretation. To evaluate Envisage, we conducted two case studies and in-depth user interviews with 14 graph analysts, The results demonstrate its effectiveness and usability in constructing, verifying, and executina complex araoh aueries.

Abstract:
Random forests are a machine learning method used to automatically classify datasets and consist of a multitude of decision trees. While these random forests often have higher performance and generalize better than a single decision tree, they are also harder to interpret. This paper presents a visualization method and system to increase interpretability of random forests. We cluster similar trees which enables users to interpret how the model performs in general without needing to analyze each individual decision tree in detail, or interpret an oversimplified summary of the full forest. To meaningfully cluster the decision trees, we introduce a new distance metric that takes into account both the decision rules as well as the predictions of a pair of decision trees. We also propose two new visualization methods that visualize both clustered and individual decision trees: (1) The Feature Plot, which visualizes the topological position of features in the decision trees, and (2) the Rule Plot, which visualizes the decision rules of the decision trees. We demonstrate the efficacy of our approach through a case study on the “Glass” dataset, which is a relatively complex standard machine learning dataset, as well as a small user study.

Abstract:
Visualizing data often entails data transformations that can reveal and hide information, operations we dub disclosure tactics. Whether designers hide information intentionally or as an implicit consequence of other design choices, tools and frameworks for visualization offer little explicit guidance on disclosure. To systematically characterize how visualizations can limit access to an underlying dataset, we contribute a content analysis of 425 examples of visualization techniques sampled from academic papers in the visualization literature, resulting in a taxonomy of disclosure tactics. Our taxonomy organizes disclosure tactics based on how they change the data representation underlying a chart, providing a systematic way to reason about design trade-offs in terms of what information is revealed, distorted, or hidden. We demonstrate the benefits of using our taxonomy by showing how it can guide reasoning in design scenarios where disclosure is a first-order consideration. Adopting disclosure as a framework for visualization research offers new perspective on authoring tools, literacy, uncertainty communication, personalization, and ethical design.

Abstract:
In data-driven storytelling contexts such as data journalism and data videos, data visualizations are often presented alongside real-world imagery to support narrative context. However, these visualizations and contextual images typically remain separated, limiting their combined narrative expressiveness and engagement. Achieving this is challenging due to the need for fine-grained alignment and creative ideation. To address this, we present SceneLoom, a Vision-Language Model (VLM)-powered system that facilitates the coordination of data visualization with real-world imagery based on narrative intents. Through a formative study, we investigated the design space of coordination relationships between data visualization and real-world scenes from the perspectives of visual alignment and semantic coherence. Guided by the derived design considerations, SceneLoom leverages VLMs to extract visual and semantic features from scene images and data visualization, and perform design mapping through a reasoning process that incorporates spatial organization, shape similarity, layout consistency, and semantic binding. The system generates a set of contextually expressive, image-driven design alternatives that achieve coherent alignments across visual, semantic, and data dimensions. Users can explore these alternatives, select preferred mappings, and further refine the design through interactive adjustments and animated transitions to support expressive data communication. A user study and an example gallery validate SceneLoom's effectiveness in inspiring creative design and facilitating design externalization.

Abstract:
Designing multiscale visualizations, particularly when the ratio between the largest scale and the smallest item is large, can be challenging, and designers have developed many approaches to overcome this challenge. We present a design space for visualization with multiple scales. The design space includes three dimensions, with eight total subdimensions. We demonstrate its descriptive power by using it to code approaches from a corpus we compiled of 52 examples, created by a mix of academics and practitioners. We demonstrate descriptive power by analyzing and partitioning these examples into four high-level strategies for designing multiscale visualizations, which are shared approaches with respect to design space dimension choices. We demonstrate generative power by analyzing missed opportunities within the corpus of examples, identified through analysis of the design space, where we note how certain examples could have benefited from different choices. We discuss patterns in the use of different dimension and strategy choices in the different visualization contexts of analysis and presentation. Supplemental materials: https://osf.io/wbrdm/ Design space website: https://marasolen.github.io/multiscale-vis-ds/

Abstract:
Although existing texture synthesis methods perform well in generating large images with irregularly repeated textures to avoid visually unrealistic repetitions, they still face significant challenges in synthesizing regular textures with densely interconnected structures. In this paper, we propose a novel neural texture synthesis method, FreNTS, which uses frequency domain information to enhance the texture synthesis process, synthesizing textures with continuous, complete, and visually realistic overall structures. The core idea is to perform the Discrete Cosine Transform on image patches to obtain the corresponding frequency domain rate information features, and then use the designed adaptive guided correspondence (AGC) loss to calculate the correlation difference between the source image and the target image in the frequency domain and spatial domains, thereby constraining the optimization of the target image to achieve high-quality texture synthesis. In addition, to better evaluate the effect of texture synthesis, we introduce Tile LPIPS as the metric for quantitative evaluation. Experimental results show that the proposed FreNTS can effectively accelerate the process of neural texture synthesis and use high-frequency information to capture better structural details to synthesize realistic textures.

Abstract:
We introduce a novel approach to simulate the interaction between fluids and thin elastic solids without any penetration. Our approach is centered around an optimization system augmented with barriers, which aims to find a configuration that ensures the absence of penetration while enforcing incompressibility for the fluids and minimizing elastic potentials for the solids. Unlike previous methods that primarily focus on velocity coherence at the fluid-solid interfaces, we demonstrate the effectiveness and flexibility of explicitly resolving positional constraints, including both explicit representation of solid positions and the implicit representation of fluid level-set interface. To preserve the volume of the fluid, we propose a simple yet efficient approach that adjusts the associated level-set values. Additionally, we develop a distance metric capable of measuring the separation between an implicitly represented surface and a Lagrangian object of arbitrary codimension. By integrating the inertia, solid elastic potential, damping, barrier potential, and fluid incompressibility within a unified system, we are able to robustly simulate a wide range of processes involving fluid interactions with lower-dimensional objects such as shells and rods. These processes include topology changes, bouncing, splashing, sliding, rolling, floating, and more.

Abstract:
Although significant progress has been made in the field of speech-driven 3D facial animation recently, the speech-driven animation of an indispensable facial component, eye gaze, has been overlooked by recent research. This is primarily due to the weak correlation between speech and eye gaze, as well as the scarcity of audio-gaze data, making it very challenging to generate 3D eye gaze motion from speech alone. In this paper, we propose a novel data-driven method which can generate diverse 3D eye gaze motions in harmony with the speech. To achieve this, we firstly construct an audio-gaze dataset that contains about 14 hours of audio-mesh sequences featuring high-quality eye gaze motion, head motion and facial motion simultaneously. The motion data is acquired by performing lightweight eye gaze fitting and face reconstruction on videos from existing audio-visual datasets. We then tailor a novel speech-to-motion translation framework in which the head motions and eye gaze motions are jointly generated from speech but are modeled in two separate latent spaces. This design stems from the physiological knowledge that the rotation range of eyeballs is less than that of head. Through mapping the speech embedding into the two latent spaces, the difficulty in modeling the weak correlation between speech and non-verbal motion is thus attenuated. Finally, our TalkingEyes, integrated with a speech-driven 3D facial motion generator, can synthesize eye gaze motion, eye blinks, head motion and facial motion collectively from speech. Qualitative and quantitative evaluations, along with a perceptual user study, demonstrate the superiority of the proposed method in generating diverse and natural 3D eye gaze motions from speech.

Abstract:
3D Gaussian Splatting has demonstrated remarkable real-time rendering capabilities and superior visual quality in novel view synthesis for static scenes. Building upon these advantages, researchers have progressively extended 3D Gaussians to dynamic scene reconstruction. Deformation field-based methods have emerged as a promising approach among various techniques. These methods maintain 3D Gaussian attributes in a canonical field and employ the deformation field to transform this field across temporal sequences. Nevertheless, these approaches frequently encounter challenges such as suboptimal rendering speeds, significant dependence on initial point clouds, and vulnerability to local optima in dim scenes. To overcome these limitations, we present FRoG, an efficient and robust framework for high-quality dynamic scene reconstruction. FRoG integrates per-Gaussian embedding with a coarse-to-fine temporal embedding strategy, accelerating rendering through the early fusion of temporal embeddings. Moreover, to enhance robustness against sparse initializations, we introduce a novel depth- and error-guided sampling strategy. This strategy populates the canonical field with new 3D Gaussians at low-deviation initial positions, significantly reducing the optimization burden on the deformation field and improving detail reconstruction in both static and dynamic regions. Furthermore, by modulating opacity variations, we mitigate the local optima problem in dim scenes, improving color fidelity. Comprehensive experimental results validate that our method achieves accelerated rendering speeds while maintaining state-of-the-art visual quality.

Abstract:
Vulnerability verification is an important process in ensuring the security of software systems. To support users in this process, we present the design study of DaV^33is, which utilizes visual event sequence analysis techniques to enable the comparison and tracing of automatically detected data flows through the software’s source code, thereby allowing users to take advantage of sequence similarities to reduce the verification workload. To that end, we characterize the domain problem based on input from domain users, describe our design rationale based on best-practices from the visual analytics literature, and evaluate individual design decisions, usability, and utility in studies with three stakeholder groups. The evaluations yielded overall positive responses, showing the suitability of our design and providing valuable insight for future research.

Abstract:
This paper challenges the convention of using graph-theoretic shortest distance in stress-based graph drawing. We propose a new paradigm based on resistance distance, derived from the graph Laplacian's spectrum, which better captures global graph structure. This approach overcomes theoretical and computational limitations of traditional methods, as resistance distance admits a natural isometric embedding in Euclidean space. Our experiments demonstrate improved neighborhood preservation and cluster faithfulness. We introduce Omega, a linear-time graph drawing algorithm that integrates a fast resistance distance embedding with random node-pair sampling for Stochastic Gradient Descent (SGD). This comprehensive random sampling strategy, enabled by efficient pre-computation of resistance distance embeddings, is more effective and robust than pivot-based sampling used in prior algorithms, consistently achieving lower and more stable stress values. The algorithm maintains O(\vert E\vert) complexity for both weighted and unweighted graphs. Our work establishes a connection between spectral graph theory and stress-based layouts, providing a practical and scalable solution for network visualization.

Abstract:
Simulating realistic fluids in virtual reality (VR) is computationally demanding, often limiting the scale and complexity of immersive environments. Existing foveated fluid simulation approaches primarily focus on spatial adaptivity. In this paper, we introduce a gaze-contingent fluid simulation system from a temporal perspective. We conduct a perceptual study to quantify the relationship between gaze eccentricity, fluid density deviation, and the perceptual threshold for simulation timesteps. Based on these findings, we fit a perceptual model that predicts the timestep requirements for maintaining perceptual realism in VR fluid animation. To exploit this model, we propose an asynchronous position-based fluids (PBF) algorithm that assigns fluid particles different local timesteps according to their visual importance and density deviation, ensuring both physical stability and perceptual validity. Our solver performs high-frequency updates in perceptually critical regions while progressively reducing updates elsewhere. A validation user study shows that our method remains perceptually indistinguishable from a high-fidelity, uniform-timestep PBF simulation. Objective evaluations further confirm that our approach improves efficiency while maintaining perceptual quality. Runtime experiments demonstrate speed-ups of up to 1.52× across diverse fluid scenarios, enabling more complex and larger-scale fluid phenomena in real-time VR. Our findings extend the paradigm of foveated fluid animation into the temporal domain, providing a perceptually grounded framework for fluid simulation in VR.

Abstract:
Translation gain is a key Redirected Walking (RDW) technique in Virtual Reality (VR) that enables users to navigate virtual environments (VEs) larger than the available physical space. The technique was originally developed to scale users' walking distance in the VE and is typically applied continuously, regardless of the user's motion state. We introduce Gait-Synced Translation Gain (GSTG), a novel approach that adapts translation gain by synchronizing it with the user's gait cycle. GSTG leverages the single-limb support phase of walking—when users are less stable and thus less sensitive to external disturbances—to apply higher levels of gain. This approach allows greater manipulation while preserving natural walking sensations and avoiding additional cybersickness. A user study comparing GSTG with continuous translation gain demonstrates significant improvements in perceived naturalness and comfort. Our results highlight the potential of gait-synchronized gain to enhance immersion, offering new possibilities for more realistic and comfortable VR locomotion.

Abstract:
Postural Instability Theory (PIT) proposes that individuals who are naturally unstable on their feet are more susceptible to cybersickness. We hypothesize that this relationship extends to locomotive VR, such that people who exhibit greater instability when walking without VR will also be more susceptible to cybersickness in a locomotive VR setup. To test this, we analyzed participants' natural walking kinematics alongside their cybersickness responses and kinematic patterns during mobile VR use. Our results showed that vertical Center of Mass movement during pre-VR walking showed promise for identifying individuals susceptible to cybersickness. Spatial stability metrics emerged as stronger predictors of cybersickness than time-series measures, suggesting that spatial characteristics of gait may be more informative indicators of susceptibility in mobile VR contexts. These findings highlight the importance of accounting for baseline postural stability when designing and personalizing mobile VR experiences.

Abstract:
We present VRGaussianAvatar, an integrated system that enables real-time full-body 3D Gaussian Splatting (3DGS) avatars in virtual reality using only head-mounted display (HMD) tracking signals. The system adopts a parallel pipeline with a VR Frontend and a GA Backend. The VR Frontend uses inverse kinematics to estimate full-body pose and streams the resulting pose along with stereo camera parameters to the backend. The GA Backend stereoscopically renders a 3DGS avatar reconstructed from a single image. To improve stereo rendering efficiency, we introduce Binocular Batching, which jointly processes left and right eye views in a single batched pass to reduce redundant computation and support high-resolution VR displays. We evaluate VRGaussianAvatar with quantitative performance tests and a within-subject user study against image- and video-based mesh avatar baselines. Results show that VRGaussianAvatar sustains interactive VR performance and yields higher perceived appearance similarity, embodiment, and plausibility. Project page and source code are available at https://vrgaussianavatar.github.io

Abstract:
In this paper, we test whether Multimodal Large Language Models (MLLMs) can match human-subject performance in tasks involving the perception of properties in network layouts. Specifically, we replicate a human-subject experiment about perceiving quality (namely stress) in network layouts using GPT-4o, Gemini-2.5 and Qwen2.5. Our experiments show that giving MLLMs the same study information as trained human participants yields performance comparable to that of human experts and exceeds that of untrained non-experts. Additionally, we show that prompt engineering that deviates from the human-subject experiment can lead to better-than-human performance in some settings. Interestingly, like human subjects, the MLLMs seem to rely on visual proxies rather than computing the actual value of stress, indicating some sense or facsimile of perception. Explanations from the models are similar to those used by the human participants (e.g., an even distribution of nodes and uniform edge lengths).

Abstract:
Effective visualization is essential for interpreting genomics data, yet researchers often face challenges in finding relevant, reusable examples. Existing tools offer limited support for searching the vast landscape of genomics visualizations, making the process of authoring new visualizations time-consuming and inefficient. To address this gap, we introduce Geranium, a data visualization retrieval system for searching and authoring genomics visualizations. Geranium supports multimodal retrieval, enabling users to query with images, text, or grammar-based specifications. Retrieved examples serve as scaffolds for authoring, providing templates that researchers can adapt with their own data, thereby streamlining the mechanics of visualization construction. Geranium integrates three embedding methods to combine specialized and general knowledge: grammar-based embeddings tailored to genomics visualizations, multimodal embeddings from a biomedical vision-language foundation model, and text embeddings from a fine-tuned large language model. For each visualization, we construct a multimodal representation that includes a Gosling specification, a pixel-based rendering, and natural language descriptions. We evaluate embedding strategies to maximize top-kk retrieval accuracy and conduct user studies with domain collaborators to gather feedback on usability. Our collection comprises 3,200 visualizations across 50 categories, ranging from single-view to coordinated multi-view designs and supporting applications from single-cell epigenomics to structural variation analysis.

Abstract:
Rule-based machine learning models, including those derived from decision trees or forests, are often considered inherently interpretable. However, human understanding is hindered by model size, rule complexity, and interdependencies between features. Moreover, rule sets extracted from ensemble models can contain contradictory, incomplete, or counterintuitive logic, even when the overall model achieves high predictive accuracy. This paper introduces a visual analytics methodology designed to support systematic exploration of rule-based model logic and its alignment with domain knowledge. Our approach integrates overview visualizations, interactive filtering, contradiction analysis, and topic modeling. This enables analysts to detect illogical or implausible rules, assess their potential impact, and refine the model to improve its interpretability and trustworthiness. A key distinction of our method is its ability to support reasoning about model behavior both with and without access to labeled data. We demonstrate the approach through two real-world case studies: evaluating logical consistency in a vessel movement classifier and analyzing feature relationships in a COVID-19 prediction model. These studies show how visual analytics can facilitate logic-focused model critique beyond traditional performance metrics and enable valuable domain-relevant insights.

Abstract:
We propose scene-adaptive strategies to efficiently allocate representation capacity for generating immersive experiences of indoor environments from incomplete observations. Indoor scenes with multiple rooms often exhibit irregular layouts with varying complexity, containing clutter, occlusion, and flat walls. We maximize the utilization of limited resources with guidance from geometric priors, which are often readily available after pre-processing stages. We record observation statistics on the estimated geometric scaffold and guide the optimal placement of bases, which greatly improves upon the uniform basis arrangements adopted by previous scalable scene representations. We also suggest scene-adaptive virtual viewpoints to compensate for geometric deficiencies inherent in view configurations in the input trajectory and impose the necessary regularization. We present a comprehensive analysis and discussion regarding rendering quality and memory requirements in several large-scale indoor scenes, demonstrating significant enhancements compared to baselines that employ regular placements.

Abstract:
Artificial facial mimicry (AFM) is increasingly used to enhance social interaction with virtual agents in immersive virtual reality. However, its psychological and ethical implications remain insufficiently explored. This article conceptualizes AFM as an effective and embodied intervention, examining the role of emotional congruence, individual differences, and clinical vulnerability in shaping user responses. We further outline methodological directions involving physiological measures and embodied coordination. By framing AFM within affective computing and embodied cognition, this work contributes to the responsible design of emotionally adaptive virtual agents.

Abstract:
Virtual reality telepresence requires immersive shared virtual environments for real-time remote collaboration across different physical scenes. It supports a wide range of applications in teleconferencing, education, and interactive simulations. However, challenges persist in identifying optimal shared virtual spaces that accommodate diverse user interaction requirements while adhering to local physical constraints during scene synthesis. In this paper, we propose an interaction-aware shared virtual scene synthesis method, which uses the large language model (LLM) to produce collaborative virtual scenes based on interaction demands from remote users in different local spaces. First, we introduce the concept of Interaction Aware Template (IAT) and its construction method using an LLM planner. Then, we propose an IAT-based affordance field alignment method for merging the local spaces of the remote users, maximally ensuring that the aligned space could support the user's desired interaction. Finally, we propose an LLM-based shared scene synthesis method according to the merged affordance field. Experiment results show that, compared to existing text-based scene synthesis and mutual space matching methods, our method achieves better Affordance Consistency, 3D Intersection over Union, and Layout Suitability on both scanned and synthesized datasets. The results of the user study demonstrate that the user's subjective perception of interaction fitness and sense of safety were significantly improved.

Abstract:
Visual generation models have achieved remarkable progress in computer graphics applications but still face significant challenges in real-world deployment. Current assessment approaches for visual generation tasks typically follow an isolated three-phase framework: test input collection, model output generation, and user assessment. These fashions suffer from fixed coverage, evolving difficulty, and data leakage risks, limiting their effectiveness in comprehensively evaluating increasingly complex generation models. To address these limitations, we propose DyEval, an LLM-powered dynamic interactive visual assessment framework that facilitates collaborative evaluation between humans and generative models for text-to-image systems. DyEval features an intuitive visual interface that enables users to interactively explore and analyze model behaviors, while adaptively generating hierarchical, fine-grained, and diverse textual inputs to continuously probe the capability boundaries of the models based on their feedback. Additionally, to provide interpretable analysis for users to further improve tested models, we develop a contextual reflection module that mines failure triggers of test inputs and reflects model potential failure patterns, supporting in-depth analysis using the logical reasoning ability of LLM. Qualitative and quantitative experiments demonstrate that DyEval can effectively help users identify max up to 2.56 timesmore generation failures than conventional methods, and uncover complex and rare failure patterns, such as issues with pronoun generation and specific cultural context generation. Our framework provides valuable insights for improving generative models and has broad implications for advancing the reliability and capabilities of visual generation systems across various domains.

Abstract:
Recent research on motion generation and text-to-motion synthesis focus on coarse-grained motion descriptions, neglecting fine-grained motion details and motion quality refinement. Additionally, current text-to-motion models, such as MotionGPT, lack multi-turn interaction capabilities, relying on single-turn and single-modality transformations, which limit their ability to integrate information from different modalities across interaction stages. These gaps leave critical questions, such as “How well is the motion performed” and “How can it be refined?” largely unaddressed. To address these issues, first, we introduce two fine-grained dance datasets—one focusing on jazz dance and the other on folk dance, which we have independently collected. Second, considering that dance motions are inherently complex and consist of long sequential actions, we introduce both global and local optimization during the motion encoding phase and employ Hidden Markov Model (HMM) temporal modeling to capture differential features between correct and incorrect movements, thereby optimizing the training process. Finally, we propose a multi-turn historical dialogue framework that enables three stages generation—motion assess, text instructions, and motion refinement—for input videos. This framework assists dance beginners by providing feedback on their movements, offering textual instructions, and delivering motion-based refinement. Experimental results on the jazz dance and folk dance datasets demonstrate that our method surpasses existing approaches in both quantitative and qualitative metrics, establishing a new benchmark for motion-text generation in the field of dance training.

Abstract:
Volume rendering techniques for scientific visualization have recently shifted toward Monte Carlo (MC) methods for their flexibility and robustness, but their use in multi-channel visualization remains underexplored. Traditional multi-channel volume rendering often relies on arbitrary, non-physically based color blending functions that hinder interpretation. We introduce multi-density Woodcock tracking, a simple extension of Woodcock tracking that leverages an MC method to produce high-fidelity, physically grounded multi-channel renderings without arbitrary blending. By generalizing Woodcock’s distance tracking, we provide a unified blending modality that also integrates blending functions from prior works. We further implement effects that enhance boundary and feature recognition. By accumulating frames in real-time, our approach delivers high-quality visualizations with perceptual benefits, demonstrated on diverse datasets.

Abstract:
Text-driven 3D editing is an emerging task that focuses on modifying scenes based on text prompts. Current methods often adapt pre-trained 2D image editors to multi-view observations, using specific strategies to combine information across views. However, these approaches still struggle with ensuring consistency across views, as they lack precise control over the sharing of information, resulting in edits with insufficient visual changes and blurry details. In this paper, we propose CoreEditor, a novel framework for consistent text-to-3D editing. At the core of our approach is a novel correspondence-constrained attention mechanism, which enforces structured interactions between corresponding pixels that are expected to remain visually consistent during the diffusion denoising process. Unlike conventional wisdom that relies solely on scene geometry, we enhance the correspondence by incorporating semantic similarity derived from the diffusion denoising process. This combined support from both geometry and semantics ensures a robust multi-view editing process. Additionally, we introduce a selective editing pipeline that enables users to choose their preferred edits from multiple candidates, creating a more flexible and user-centered 3D editing process. Extensive experiments demonstrate the effectiveness of CoreEditor, showing its ability to generate high-quality 3D edits, significantly outperforming existing methods.

Abstract:
Point cloud denoising is a fundamental yet challenging task in computer graphics. Existing solutions typically rely on supervised training on synthesized noise. However, real-world noise often exhibits greater complexity, causing learning-based methods trained on synthetic noise to struggle when encountering unseen noise–a phenomenon we refer to as noise misalignment. To address this challenge, we propose LaPDA (Latent-space Point cloud Denoising with Adaptivity), a neural network explicitly designed to mitigate noise misalignment and enhance denoising robustness. LaPDA consists of two key stages. First, we adaptively model noise in the latent space, aligning unseen noise distributions with the known training distributions or adjusting them toward distributions with lower noise scales. Training objectives at this stage are formulated based on controlled synthetic noise with varying intensity levels. Second, we introduce a gradual noise removal module that optimizes the spatial distribution of the adaptively adjusted noisy points. Extensive experiments conducted on both synthetic and scanned datasets demonstrate that LaPDA achieves enhanced accuracy and robustness compared to state-of-the-art methods.

Abstract:
Zero-shot text-to-video diffusion models are crafted to expand pre-trained image diffusion models to the video domain without additional training. In recent times, prevailing techniques commonly rely on existing shapes as constraints and introduce inter-frame attention to ensure texture consistency. However, such shape constraints tend to restrict the stylized geometric deformation of videos and inadvertently neglect the original texture characteristics. Furthermore, existing methods suffer from flickering and inconsistent facial expressions. In this paper, we present DiffPortraitVideo. The framework employs a diffusion model-based feature and attention injection mechanism to generate key frames, with cross-frame constraints to enforce coherence and adaptive feature fusion to ensure expression consistency. Our approach achieves high spatio-temporal and expression consistency while retaining the textual and original image properties. Extensive and comprehensive experiments are conducted to validate the efficacy of our proposed framework in generating personalized, high-quality, and coherent videos. This not only showcases the superiority of our method over existing approaches but also paves the way for further research and development in the field of text-to-video generation with enhanced personalization and quality.

Abstract:
Surface mesh offsetting is a fundamental operation in various applications (e.g., shape modeling). Implicit methods that contour a volumetric distance field are robust at handling intersection defects, but it is challenging to apply shape control (e.g., preserving sharp features in the input shape) and to avoid undesired topology changes. Explicit methods, which move vertices towards the offset surface (with possible adaptivity), can address the above issues, but it is hard to avoid intersection issues. To combine the advantages of both, we propose a variational framework that takes mesh vertex locations as variables while simultaneously involving a smooth winding-number field associated with the mesh. Under various shape regularizations (e.g., sharp feature preservation) formulated on the mesh, the objective function mainly requires that the input mesh lie on the offset contour of the field induced by the resulting mesh. Such a combination inherits the ability to apply flexible shape regularizations from explicit methods and significantly alleviates intersection issues because of the field. Moreover, the optimization problem is numerically friendly by virtue of the differentiability of the field w.r.t. the mesh vertices. Results show that we can offset a mesh while preserving sharp features of the original surface, restricting selected parts to quadric surfaces and penalizing intersections.

Abstract:
Accurate illumination is crucial for many imaging and vision applications, and skies are the dominant source of lighting in many scenes. Most existing work for representing sky illumination has focused on clear skies or more recently generative approaches for synthesizing clouds. However, these are very limited in that they assume distant illumination and do not capture the 3D properties of clouds. This paper presents a novel and principled approach to extract 3D whole-sky volumetric representations of clouds which can be used for imaging applications. Our approach extracts clouds from a single fisheye capture of the sky via an iterative optimization process. We achieve this by exploiting the physical properties of light scattering in clouds and use these to drive a domain-specific light transport simulation algorithm to render the images required for optimization. Results for this method provide high accuracy when re-rendering with our reconstructed clouds compared to real captures, and also enable novel uses of environment maps such as inclusion of captured clouds in renderings, cloud shadows, and more accurate aerial perspective and lighting.

Abstract:
3D reconstruction is a critical technology with significant implications for applications such as urban planning, autonomous driving, and virtual reality. Recent advances in 3D Gaussian Splatting (3DGS) have demonstrated impressive results in small-scale scenes, achieving high-quality reconstructions with real-time rendering capabilities. However, when applied to large-scale scenes, existing 3DGS methods face significant challenges due to the exponential growth of model size, often exceeding the memory capacity of consumer-grade GPUs and making training and rendering infeasible. In this paper, we propose a structure-guided memory-efficient 3DGS framework that uses only half the memory of current large-scale 3DGS methods while maintaining state-of-the-art reconstruction accuracy. Specifically, we introduce a structure-guided density control mechanism that uses a heuristic approach to split Gaussian ellipsoids in challenging regions and optimizes their attributes during densification, significantly reducing memory storage requirements while preserving structural details with fewer ellipsoids. Moreover, we propose a novel structure loss to supervise the learning of scene structural information, enabling the model to better capture and preserve geometric details such as straight lines and edges, further enhancing reconstruction accuracy. We also propose the largest known drone dataset for 3D reconstruction, comprising over 10,000 high-resolution images covering more than 2.5 million square meters. Extensive experiments on multiple benchmark datasets and our proposed dataset demonstrate that our new method is highly memory-efficient with high accuracy. We strongly recommend you to watch our demo at https://lvzinan.github.io/STGS.github.io/.

Abstract:
Designing freeform surfaces to control light based on real-world illumination patterns is challenging, as existing caustic lens designs often assume oversimplified point or parallel light sources. We propose representing surface light sources using an optimized set of point sources, whose parameters are fitted to the real light source’s illumination using a novel differentiable rendering framework. Our physically-based rendering approach simulates light transmission using flux, without requiring prior knowledge of the light source’s intensity distribution. To efficiently explore the light source parameter space during optimization, we apply a contraction mapping that converts the constrained problem into an unconstrained one. Using the optimized light source model, we then design the freeform lens shape considering flux consistency and normal integrability. Simulations and physical experiments show our method more accurately represents real surface light sources compared to point-source approximations, yielding caustic lenses that produce images closely matching the target light distributions.

Abstract:
Urban visual querying leverages visual representations and interactions to depict the domain of interest and express related requests for exploring complex datasets, which is usually an iterative process. One main challenge of this process is the vast search space in terms of identifying querying conditions, observing querying results, and making the subsequent queries. This paper proposes a novel acceleration scheme that intelligently recommends a small set of querying results subject to previous queries. Central to our approach is a reinforcement learning based approach that trains a recommendation agent by simulating user behavior and characterizing the search space. We propose a mixed-initiative urban visual query scheme to enhance the exploration process additionally. We evaluate our approach by performing qualitative and quantitative experiments on a real-world scenario. The experimental results demonstrate the capability of reducing user workload, achieving optimized querying, and improving analysis efficiency.

Affiliations: State Key Laboratory of Tibetan Intelligent Information Processing and Application, School of Artificial Intelligence, Nanjing University of Posts and Telecommunications, Nanjing, China; State Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University, Beijing, China; School of Computer Science and Technology, China University of Mining and Technology, Jiangsu, China; School of Computer Science, Nanjing University of Posts and Telecommunications, Jiangsu, China

Abstract:
3D multi-object arrangement is a fundamental task in VR that relies on accurate and natural initial selection alongside rapid and convenient subsequent manipulation to ensure high efficiency. However, existing methods fail to support efficient multi-object arrangement in highly occluded scenes with densely packed candidate objects through controller-free natural interactions. In this article, we propose an efficient, scene-aware multi-object arrangement method (MOA) designed for fast, precise, and convenient object arrangement. First, MOA introduces an importance-driven multi-object initial selection algorithm that assigns higher spatiotemporally correlated object importance (IMP) to target objects, establishing a natural multi-object initial selection mode that enables quick and accurate selection of high-IMP objects. Subsequently, it presents an auxiliary-structure-guided multi-object manipulation algorithm that constructs an auxiliary manipulation structure to assist subsequent multi-object manipulation, alongside a multi-modal interaction mode that facilitates swift and natural manipulation. Compared to state-of-the-art controller-free and controller-based methods, MOA significantly improves task performance, reduces task load, and enhances convenience in complex multi-object arrangement scenes involving hundreds of highly occluded objects need to be arranged.

Abstract:
We propose a semantic-aware neural reconstruction method to generate 3D high-fidelity models from sparse images. To tackle the challenge of severe radiance ambiguity caused by mismatched features in sparse input, we enrich neural implicit representations by adding patch-based semantic logits that are optimized together with the signed distance field and the radiance field. A novel regularization based on the geometric primitive masks is introduced to mitigate shape ambiguity. The performance of our approach has been verified in experimental evaluation. The average chamfer distances of our reconstruction on the DTU dataset can be reduced by 44% for SparseNeuS and 20% for VolRecon. When working as a plugin for those dense reconstruction baselines such as NeuS and Neuralangelo, the average error on the DTU dataset can be reduced by 69% and 68% respectively.

Abstract:
The security of individual privacy is paramount for trajectory publication, while preserving trajectory utility is also essential to serve analysis tasks such as urban planning and transportation development. To assess and maintain trajectory utility, existing studies consider geographic context. However, innate semantic characteristics of trajectories (e.g., origin, destination, stay point, path) have been overlooked, which prevents data owners from specifying task-specific utility measurements and, consequently, from achieving a delicate balance between privacy and utility. This paper proposes an interactive trajectory publishing approach driven by flexible utility considerations, which processes trajectory points according to their semantics to fulfill diverse utility requirements. Concretely, we decouple trajectories into origin–destination (OD) and path components: ODs are generalized into regions to satisfy kk-anonymity, and paths are sanitized within each OD group using road-network–aware differential privacy under predefined privacy constraints. We also develop a visual interface to support exploration and comprehension of privacy-preserving solutions, through which we incorporate human knowledge into the privacy scheme. Experiments on real-world urban datasets demonstrate the effectiveness of our approach.

Abstract:
Despite the widespread use of Uniform Manifold Approximation and Projection (UMAP), the impact of its stochastic optimization process on the results remains underexplored. We observed that it often produces unstable results where the projections of data points are determined mostly by chance rather than reflecting neighboring structures. To address this limitation, we introduce (r,d)-stability to UMAP: a framework that analyzes the stochastic positioning of data points in the projection space. To assess how stochastic elements—specifically, initial projection positions and negative sampling—impact UMAP results, we introduce “ghosts“, or duplicates of data points representing potential positional variations due to stochasticity. We define a data point's projection as (r,d)-stable if its ghosts perturbed within a circle of radius r in the initial projection remain confined within a circle of radius d for their final positions. To efficiently compute the ghost projections, we develop an adaptive dropping scheme that reduces a runtime up to 60% compared to an unoptimized baseline while maintaining approximately 90% of unstable points. We also present a visualization tool that supports the interactive exploration of the (r,d)-stability of data points. Finally, we demonstrate the effectiveness of our framework by examining the stability of projections of real-world datasets and present usage guidelines for the effective use of our framework.

Abstract:
Connectomics, a subfield of neuroscience, aims to map and analyze synapse-level wiring diagrams of the nervous system. While recent advances in deep learning have accelerated automated neuron and synapse segmentation, reconstructing accurate connectomes still demands extensive human proofreading to correct segmentation errors. We present SynAnno, an interactive tool designed to streamline and enhance the proofreading of synaptic annotations in large-scale connectomics datasets. SynAnno integrates into existing neuroscience workflows by enabling guided, neuron-centric proofreading. To address the challenges posed by the complex spatial branching of neurons, it introduces a structured workflow with an optimized traversal path and a 3D mini-map for tracking progress. In addition, SynAnno incorporates fine-tuned machine learning models to assist with error detection and correction, reducing the manual burden and increasing proofreading efficiency. We evaluate SynAnno through a user and case study involving seven neuroscience experts. Results show that SynAnno significantly accelerates synapse proofreading while reducing cognitive load and annotation errors through structured guidance and visualization support. The source code and interactive demo are available at: https://github.com/PytorchConnectomics/SynAnno.

Abstract:
Visualization knowledge bases enable computational reasoning and recommendation over a visualization design space. These systems evaluate design trade-offs using numeric weights assigned to different features (e.g., binning a variable). Feature weights can be learned automatically by fitting a model to a collection of chart pairs, in which one chart is deemed preferable to the other. To date, labeled chart pairs have been drawn from published empirical research results; however, such pairs are not comprehensive, resulting in a training corpus that lacks many design variants and fails to systematically assess potential trade-offs. To improve knowledge base coverage and accuracy, we contribute data augmentation techniques for generating and labeling chart pairs. We present methods to generate novel chart pairs based on design permutations and by identifying under-assessed features-leading to an expanded corpus with thousands of new chart pairs, now in need of labels. Accordingly, we next compare varied methods to scale labeling efforts to annotate chart pairs, in order to learn updated feature weights. We evaluate our methods in the context of the Draco knowledge base, demonstrating improvements to both feature coverage and chart recommendation performance.

Abstract:
Research on affective visualization design has shown that color is an especially powerful feature for influencing the emotional connotation of visualizations. Associations between colors and emotions are largely driven by lightness (e.g., lighter colors are associated with positive emotions, whereas darker colors are associated with negative emotions). Designing visualizations to have all light or all dark colors to convey particular emotions may work well for visualizations in which colors represent categories and spatial channels encode data values. However, this approach poses a problem for visualizations that use color to represent spatial patterns in data (e.g., colormap data visualizations) because lightness contrast is needed to reveal fine details in spatial structure. In this study, we found it is possible to design colormaps that have strong lightness contrast to support spatial vision while communicating clear affective connotation. We also found that affective connotation depended not only on the color scales used to construct the colormaps, but also the frequency with which colors appeared in the map, as determined by the underlying dataset (data-dependence hypothesis). These results emphasize the importance of data-aware design, which accounts for not only the design features that encode data (e.g., colors, shapes, textures), but also how those design features are instantiated in a visualization, given the properties of the data.

Abstract:
We present PiCCL (Pictorial Chart Composition Language), a new language that enables users to easily create pictorial charts using a set of simple operators. To support systematic construction while addressing the main challenge of expressive pictorial chart authoring-manual composition and fine-tuning of visual properties-PiCCL introduces a parametric representation that integrates data-driven chart generation with graphical composition. It also employs a lazy data-binding mechanism that automatically synthesizes charts. PiCCL is grounded in a comprehensive analysis of real-world pictorial chart examples. We describe PiCCL's design and its implementation as piccl.js, a JavaScript-based library. To evaluate PiCCL, we showcase a gallery that demonstrates its expressiveness and report findings from a user study assessing the usability of piccl.js. We conclude with a discussion of PiCCL's limitations and potential, as well as future research directions.

Abstract:
Tens of thousands of people have represented data by creating data-encoding textile pieces like blankets, scarves, and more. A prototypical example is the temperature blanket, which represents the weather through rows or blocks of different colors mapped to temperature ranges. While researchers have used fiber arts mediums to create exploratory projects, data visualization and physicalization research has largely not engaged with examples from this enormous and diverse community. We explore the space of data textiles, or fiber arts that encode information, by surveying creators (i.e., data fiber artists) on their projects and processes. We create a corpus of 159 examples of data textiles and present a schema characterizing the data encoding methods used in these projects. We also gather insights into creators' data workflows as well as their motives and discoveries through making with their data. Creators of data textiles use distinct processes to map their data, building fabric from component structures and substructures while using material properties like color and texture. From diverse data-tracking procedures, creators use and relate to data in varied ways. Working on these pieces also contributes to the creators' personal growth and data understanding. Our findings point to new opportunities for visualization, including opportunities to support fiber artists with tools formatted to their needs and opportunities to incorporate concepts from data textiles into other types of visualization (e.g., using texture, structural layouts, colorways).

Abstract:
This application paper investigates the stability of hydrogen bonds (H-bonds), as characterized by the Quantum Theory of Atoms in Molecules (QTAIM). First, we contribute a database of 4544 electron densities associated to four isomers of water hexamers (the so-called Ring, Book, Cage and Prism), generated by distorting their equilibrium geometry under various structural perturbations, modeling the natural dynamic behavior of molecular systems. Second, we present a new stability measure, called bond occurrence rate, associating each bond path present at equilibrium with its rate of occurrence within the input ensemble. We also provide an algorithm, called BondMatcher, for its automatic computation, based on a tailored, geometry-aware partial isomorphism estimation between the extremum graphs of the considered electron densities. Our new stability measure allows for the automatic identification of densities lacking H-bond paths, enabling further visual inspections. Specifically, the topological analysis enabled by our framework corroborates experimental observations and provides refined geometrical criteria for characterizing the disappearance of H-bond paths. Our electron density database and our C++ implementation are available at this address: https://github.com/thom-dani/BondMatcher.

Abstract:
Structural analysis is essential in modern industrial design, where engineers iteratively refine geometry models based on stress simulations to achieve optimized designs. However, comparing stress distributions across multiple model variants remains challenging due to the complexity of stress fields, which are high-dimensional, unevenly distributed, and dependent on intricate geometric structures. Existing tools primarily support single-model analysis and lack dedicated functionalities for multi-model comparison. As a result, engineers must rely on manual, cognitively demanding visual inspections, making it difficult to systematically identify and interpret stress variations across design iterations. To address these limitations, we propose StressDiffVis, a visual analytics approach that facilitates stress field comparison across multiple structural models. StressDiffVis employs a volumetric representation to encode stress distributions while minimizing occlusion, enabling voxel-wise difference analysis for model comparison. To support localized analysis, we introduce model segmentation, grouping voxels with similar stress patterns across models. StressDiffVis integrates these techniques into an interactive interface with a tree view, organizing models by the iterative design process, and a comparison view, using a matrix layout for detailed comparisons. We demonstrate the effectiveness of StressDiffVis through two case studies illustrating its utility in comparative stress analysis. In addition, expert interviews confirm its potential to enhance engineering workflows.

Abstract:
What impressions might readers form with visualizations that go beyond the data they encode? In this paper, we build on recent work that demonstrates the socio-indexical function of visualization, showing that visualizations communicate more than the data they explicitly encode. Bridging this with prior work examining public discourse about visualizations, we contribute an analytic framework for describing inferences about an artifact's social provenance. Via a series of attribution-elicitation surveys, we offer descriptive evidence that these social inferences: (1) can be studied asynchronously, (2) are not unique to a particular sociocultural group or a function of limited data literacy, and (3) may influence assessments of trust. Further, we demonstrate (4) how design features act in concert with the topic and underlying messages of an artifact's data to give rise to such ‘beyond-data’ readings. We conclude by discussing the design and research implications of inferences about social provenance, and why we believe broadening the scope of research on human factors in visualization to include sociocultural phenomena can yield actionable design recommendations to address urgent challenges in public data communication.

Abstract:
The apparel industry has witnessed a significant shift toward 3D virtual design, driven by advances in simulation and digitization technologies. Central to this transformation are digital twins, which allow designers to test and refine garment designs virtually, reducing reliance on physical prototyping. A critical aspect of this process is accurately capturing the drape behavior of fabrics, as it impacts both the aesthetic and functional properties of garments. Traditional fabric digitization methods involve mechanical testing of fabric samples, but they are labor-intensive, costly, and require specialized equipment. Recently, AI-based digitization techniques have emerged, offering potential advantages in terms of cost and logistics, though their accuracy remains uncertain. This study addresses the gap in knowledge regarding the accuracy of AI-based fabric digitization by utilizing the Cusick drape test—a standardized method for assessing fabric drape behavior. We introduce a reference digitization pipeline based on traditional mechanical testing and perform a comprehensive comparative study of six commercially available digitization methods, including both traditional and AI-based approaches. Using a diverse set of fabrics, we evaluate the accuracy of the digitized fabrics against real-world measurements. Our results demonstrate that AI-based methods can achieve competitive accuracy, while also highlighting areas for further improvement. We present a publicly available dataset containing real fabric properties, their digitized versions, as well as their corresponding real and digital drape metrics, which can serve as a valuable resource for further research and development in fabric digitization.

Abstract:
We propose an efficient method to compute a small set of integer-constrained cone singularities, which induce a rotationally seamless conformal parameterization with low distortion. Since the problem only involves discrete variables, i.e., vertex-constrained positions, integer-constrained angles, and the number of cones, we alternately optimize these three types of variables to achieve tractable convergence. Central to high efficiency is an explicit construction algorithm that reduces the optimization problem scale to be slightly greater than the number of integer variables for determining the optimal angles with fixed positions and numbers, even for high-genus surfaces. In addition, we derive a new derivative formula that allows us to move the cones, effectively reducing distortion until convergence. Combined with other strategies, including repositioning and adding cones to decrease distortion, adaptively selecting a constrained number of integer variables for efficient optimization, and pairing cones to reduce the number, we quickly achieve a favorable tradeoff between the number of cones and the parameterization distortion. We demonstrate the effectiveness and practicability of our cones by using them to generate rotationally seamless and low-distortion parameterizations on a massive test data set. Our method demonstrates an order-of-magnitude speedup (30× faster on average) compared to state-of-the-art approaches while maintaining comparable cone numbers and parameterization distortion.

Abstract:
Numerous diffusion models have been developed for 2D image synthesis and editing, and recently they are extended to 3D scene editing tasks. However, editing 3D scenes is still in its early stages, and the challenges of scene representations and multi-view consistency need to be addressed. A notable limitation of existing approaches is the need for specific modules for different edits and model retraining for each scene. To tackle these issues, we propose a novel and versatile text-driven 3D scene editing method, termed DN2N, which allows for the direct acquisition of the editing results without the requirement for retraining. Our method employs off-the-shelf text-based editing models of 2D images to modify the multi-view images of a 3D scene. A content filtering process is then applied to discard poorly edited images that disrupt 3D consistency. We consider the remaining inconsistency as a problem of removing noise perturbations and solve it by generating data with similar perturbation characteristics for training. We develop a versatile NeRF model structure and propose two novel cross-view regularization terms to help the DN2N mitigate these perturbations. Empirical results show that our method achieves multiple editing types based solely on text prompts, including but not limited to appearance editing, weather transition, object changing, and style transfer. Most importantly, DN2N exhibits a versatility of editing capabilities, eliminating the need to customize or retrain editing models for specific scenes or editing types. Namely, DN2N achieves comparable total editing time to the 3DGS-based editing method, enhancing its practical value.

Abstract:
Deep neural networks (DNNs) achieve state-of-the-art performance in many vision tasks, yet understanding their internal behavior remains challenging, particularly how different layers and activation channels contribute to class separability. We introduce ChannelExplorer, an interactive visual analytics tool for analyzing image-based outputs across model layers, emphasizing data-driven insights over architecture analysis for exploring class separability. ChannelExplorer begins with a dataset-level overview and progressively drills down to individual examples, summarizing activations across layers along the way. It presents these results primarily through three coordinated views: a Scatterplot View to reveal inter and intra-class confusion, a Jaccard Similarity View to quantify activation overlap, and a Heatmap View to inspect activation channel patterns. Our technique supports diverse model architectures, including CNNs, GANs, ResNet, and Stable Diffusion models. We demonstrate the capabilities of ChannelExplorer through four use-case scenarios: (1) generating class hierarchy in ImageNet, (2) finding mislabeled images, (3) identifying activation channel contributions, and (4) locating latent states’ position in the Stable Diffusion model. Finally, we evaluate the tool with expert users.

Abstract:
Corporate financial news is pivotal for market decisions, but often overwhelms general audiences. While data videos effectively bridge this comprehension gap, their production remains a bottleneck for journalists. We present NewsVis, an authoring tool powered by Generative Artificial Intelligence (GenAI) that automates the transformation of unstructured narratives and raw financial datasets into professional data videos. Unlike generic models, our pipeline ensures factual accuracy through a domain-specific taxonomy of financial attributes and optimizes visual information presentation via a multimodal layout algorithm. Additionally, a human-in-the-loop interface empowers journalists to audit and calibrate generative outputs. Comprehensive quantitative and qualitative evaluations demonstrate that NewsVis significantly reduces production barriers while enhancing information accessibility for viewers.

Abstract:
Real-time global illumination rendered with low variance remains a persistent challenge. Many engines employ irradiance probes as a relatively cheap technique, but constrained computational budgets often lead to flickering artifacts in rendered images. In this paper, we propose spatial multiple importance sampling, which reuses ray–surface intersection data among real-time irradiance probes to significantly reduce flicker caused by variance, enabling efficient computation in complex scenes under limited ray tracing budgets. Moreover, our approach incorporates a probe selection mechanism to enhance reuse efficiency and a visibility estimation method to mitigate bias. Experimental results demonstrate that our method significantly reduces variance at a fixed ray tracing cost, delivering high-quality, stable outputs in real-time scenarios.

Abstract:
This paper presents IMAGGarment, a fine-grained garment generation (FGG) framework that enables high-fidelity garment synthesis with precise control over silhouette, color, and logo placement. Unlike existing methods that are limited to single-condition inputs, IMAGGarment addresses the challenges of multi-conditional controllability in personalized fashion design and digital apparel applications. Specifically, IMAGGarment employs a two-stage training strategy to separately model global appearance and local details, while enabling unified and controllable generation through end-to-end inference. In the first stage, we propose a global appearance model that jointly encodes silhouette and color using a mixed attention module and a color adapter. In the second stage, we present a local enhancement model with an adaptive appearance-aware module to inject user-defined logos and spatial constraints, enabling accurate placement and visual consistency. To support this task, we release GarmentBench, a large-scale dataset comprising over 180 K garment samples paired with multi-level design conditions, including sketches, color references, logo placements, and textual prompts. Extensive experiments demonstrate that our method outperforms existing baselines, achieving superior structural stability, color fidelity, and local controllability performance.

Abstract:
This paper presents a visual analytics workflow for detecting stable cross-impact patterns in time series pairs. A sliding window technique computes multiple impact measures, including a novel Kendall’s tau variant that tolerates minor fluctuations. Evaluating these measures across various time lags reveals dynamic relationships between time series. An interactive Ikat plot facilitates the exploration of impact distributions, helping identify intervals where specific cross-impacts remain stable (e.g., trends in one series followed by similar or opposite trends in another after a lag). These intervals are extracted as events, whose temporal (and, when applicable, spatial) distributions can be analyzed to uncover broader patterns across multiple time series pairs and over extended time spans. This includes identifying co-occurring cross-impacts and variations in cross-impact presence or type across different periods and data subsets. Experiments on real-world datasets demonstrate the framework’s ability to isolate robust patterns, providing a scalable and interpretable approach to analyzing complex temporal dynamics.

Abstract:
Reconstructing editable CAD models from mesh data remains a fundamental yet challenging problem in reverse engineering. Existing methods often struggle to achieve both geometric accuracy and topological consistency, especially when recovering analytic design intent from complex meshes. We present an interactive CAD reverse modeling framework based on variational curve approximation, which bridges mesh segmentation and parametric reconstruction in a unified manner. Our method extracts surface-aligned cutting curves from segmented regions and refines them through a variational approximation process that encodes primitive attributes and geometric constraints. This enables the automatic recovery of closed loops, constraint relations, and operation parameters, leading to a compact and topologically consistent CAD representation. In addition, a coplanar profile detection and voxel-similarity–based Boolean inference are developed to restore the topological order of the original modeling sequence. Extensive experiments demonstrate that our approach achieves higher geometric fidelity, better structural consistency, and improved automation compared with state-of-the-art techniques. The reconstructed models are watertight, topologically consistent, and readily applicable to downstream CAD design tasks.

Abstract:
Visualization design is often a demanding process that involves trying different encodings, exploring different data transformations, and refining details. While there have been important studies of these workflows, the lower-level, code-intensive practices remain underexplored. Exploratory notebook tools have allowed designers to rapidly iterate on visualizations. We use publicly-available version histories of notebooks to study how users work in these environments, observing both how they build new visualizations from existing templates or previous work and how they refine visualizations over time. We examine the interplay between data manipulation and visualization, and classify the types of changes made when updating visual encodings. We also analyze the impact of different frameworks by comparing two code-oriented libraries and a chart wizard. Finally, we examine how interactions with notebooks have changed over the years, including after the widespread availability of AI. These analyses help us understand how users iterate to produce visualizations over time using different frameworks.

Abstract:
Bias in machine learning datasets occurs when certain attributes are unfairly associated, e.g., serious males being mostly linked with law enforcement officers in job-related image datasets. Training models on biased datasets will degrade model performance and lead to fairness issues, particularly for underrepresented groups. Existing bias detection methods mainly focus on explicit biases associated with predefined attributes (e.g., gender and ethnicity) while overlooking implicit biases associated with subtler, dataset-specific attributes (e.g., facial expressions and attire). To address this gap, we present BiasField, an interactive tool that offers a closed-loop workflow for detecting, analyzing, and mitigating bias. Central to BiasField is the adaptive detection of both explicit and implicit biases, a process facilitated by the automatic extraction of the dataset-specific attributes. It then employs a plant-growth metaphor to visualize these biases, enabling structured analysis to identify similar biases and track how they strengthen with additional attributes. Finally, confirmed biases are mitigated through targeted generative data augmentation. A user study, two case studies, and an expert study are conducted to demonstrate its capability to detect, analyze, and mitigate complex biases.

Abstract:
We propose Probabilistic Inclusion Depth (PID) for the ensemble visualization of scalar fields. By introducing a probabilistic inclusion operator \subset_p, our method is a general data depth model supporting ensembles of fuzzy contours, such as soft masks from modern segmentation methods, and conventional ensembles of binary contours. We also advocate for extending contour extraction in scalar field ensembles to become a fuzzy decision by considering the probabilistic distribution of an isovalue to encode the sensitivity information. To reduce the complexity of the data depth computation, an efficient approximation using the mean probabilistic contour is devised. Furthermore, an order-of-magnitude reduction in computational time is achieved with an efficient parallel algorithm on the GPU. Our new method enables the computation of contour boxplots for ensembles of probabilistic masks, ensembles defined on various types of grids, and large 3D ensembles not studied by existing methods. The effectiveness of our method is evaluated through numerical comparisons with existing techniques on synthetic datasets, examples of real-world ensemble datasets, and expert feedback.

Abstract:
The course of history is inseparable from the actions of historical figures. From the perspective of contemporaneity, an important challenge is how to visualize large numbers of historical figures and their dynamic associations on a unified temporal scale. Yet existing large-scale data solutions often rely on aggregation strategies that diminish the visibility of key figures, while failing to provide smooth contextual transitions between overview and detail. Moreover, current approaches to representing inter-figure associations lack effective integration of the surrounding historical context. To address these challenges, we propose TimeScape, a large-scale biographical data exploration system supporting multi-resolution analysis. Anchored in absolute time, the system highlights the evolving dynamics of historical development. Through multi-level layouts, sampling, and semantic zooming, it enables efficient navigation of large-scale data, allowing users to move seamlessly between overview and detail and to “wander” freely across the historical landscape. The system further allows iterative switching of focal figures to explore concrete inter-personal associations, aligning them on the temporal axis so that these connections are grounded in traceable evidence and explicit context. Case studies and expert interviews demonstrate that TimeScape effectively supports multi-perspective understanding and exploration of large-scale biographical data, offering historians a new visualization pathway for research.

Abstract:
We propose event-based vibrotactile feedback as an effective approach for enhancing bare-hand XR interaction, delivered through commodity smartwatches that users already wear. While controllers and haptic gloves provide rich tactile feedback, they introduce additional hardware that users must carry and wear throughout the day. Importantly, they interpose hardware between the user and the virtual world, occupying the hands, constraining finger motion, and undermining truly unobstructed bare-hand interaction. Our proposed approach delivers short discrete pulses at key manipulation moments such as object contact and state changes, providing tactile confirmation at the wrist without requiring additional hardware or disrupting finger movement. Two user studies with 26 participants demonstrate that this referred feedback significantly enhances user experience at either wrist placement, with 94% finding it helpful and particular benefits during complex tasks like knob rotation where haptic cues reduced visual attention demands. These findings establish design guidelines for integrating accessible haptics into everyday XR through personal devices, supporting broader adoption of natural hand interaction.

Abstract:
Structured light (SL) is a popular approach for 3D reconstruction. Most SL techniques rely on frame-based cameras and are often not robust in high-speed dynamic scenes. Recently, event cameras have sparked growing interest in high-speed SL imaging, due to their high temporal resolution. The event-based SL enjoys the high-speed data acquisition, however, most existing methods tend to pursue the reconstruction accuracy but sacrificing the computational efficiency, limiting the applicability in real-world scenarios. To this end, we propose E2SL, an Efficient deep network tailored for monocular Event-based SL. Specifically, E2SL comprises three key components: binary embedding lookup table (BE-LUT), spatial context enhancement (SCE), and geometric-prior regression (GPR). Given the input event frame, BE-LUT, which is precomputed and stored, first retrieves the features efficiently. Then, SCE extends the receptive field of the features and captures the spatial context. Finally, GPR conducts the geometric-prior-based tree classification for fast and robust depth estimation. To support training and evaluation, we contribute an event-based SL simulator, which generates a large-scale and diverse synthetic dataset. Besides, we develop an event-based SL prototype and collect a dataset with accurate ground truth for real-world evaluation. Extensive experiments demonstrate that our method achieves state-of-the-art accuracy while maintaining a per-frame reconstruction time of 7.7 ms, meeting the demands of high-speed depth sensing. The code and dataset are available on the project page https://dongxin000.github.io/E2SL/.

Abstract:
Immersive VR applications demand high resolutions and refresh rates, posing significant challenges for real-time rendering. Foveated rendering mitigates this cost by exploiting properties of the Human Visual System (HVS), but conventional approaches often rely on oversimplified heuristic models that neglect high-level attentional cues, resulting in artifacts in peripheral regions. To this end, we present a neural saliency-driven foveated ray tracing framework that overcomes these limitations. Our method introduces a motion-aware foveation model to capture temporal dynamics and employs a lightweight convolutional neural network to predict saliency maps that reflect complex attentional patterns derived from eye-gaze data. The combination of these guides adaptive path tracing and filtering, enabling perceptually optimized rendering with minimal artifacts. Experimental results show that our approach improves perceptual quality over prior methods while sustaining real-time performance.

Abstract:
Automating immersive VR scene creation remains a primary research challenge. Existing methods typically rely on complex geometry with post-simplification, resulting in inefficient pi pelines or li mited re alism. In th is paper, we introduce Im merseGen, a novel agent-guided framework for compact and photorealistic world generation that decouples realism from exhaustive geometric modeling. ImmerseGen represents scenes as hierarchical compositions of lightweight geometric proxies with synthesized RGBA textures, facilitating real-time rendering on mobile VR headsets. We propose terrain-conditioned texturing for base world generation, combined with context-aware texturing for scenery, to produce diverse and visually coherent worlds. VLM-based agents employ semantic grid-based analysis for precise asset placement and enrich scenes with multimodal enhancements such as visual dynamics and ambient sound. Experiments and real-time VR applications demonstrate that ImmerseGen achieves superior photorealism, spatial coherence, and rendering efficiency compared to existing methods.

Abstract:
Intuitive motion control is essential for virtual reality, allowing users to manipulate objects naturally while receiving realistic and responsive visual feedback. 3D Gaussian splatting provides real-time, photorealistic scene rendering, making it promising for virtual reality applications. Still, it falls short in accurate motion control of dynamic objects due to its unstructured global motion representation and redundant motion learning. To address these problems, we propose a motion hierarchical Gaussian based dynamic control method. First, a motion hierarchical Gaussian representation is introduced and initialized with semantic and deformation information. Then a motion hierarchical decomposition method is proposed to optimize the local motion in the representation. The representation is next optimized by a local motion analysis based refinement method. We also design a set of motion control operations for the motion hierarchical Gaussian. Experimental results show that our method achieves high-precision motion reconstruction, accurate motion decomposition, real-time, intuitively and immersive VR motion control.

Abstract:
Design spaces serve as a conceptual framework that enables designers to explore feasible solutions, yet their lack of computational formalization limits integration with AI-driven design automation. To address this, we introduce a structured design space model that formalizes design spaces with orthogonal dimensions and discrete elements, making them machine-interpretable and executable. Building on this model, we present IDEA, a fully automated design exploration framework to generate effective outcomes based on user requirements and design spaces. Specifically, IDEA leverages large language models (LLMs) for constraint generation, incorporates a constraint-guided Monte Carlo Tree Search (MCTS) algorithm to explore the space, and instantiates abstract decisions into domain-specific implementations. We evaluate IDEA in two design scenarios: data-driven article and standard visualization, supported by comparativeratings, expert interviews, and quantitative experiments. Results demonstrate the IDEA’s adaptability across domains and its capability to produce high-quality designs.

Abstract:
Scientific applications are generating unprecedented volumes of data that overwhelm storage and transmission systems, posing significant challenges for the design of data management tools and scientific databases. Lossy compression has emerged as a promising strategy to address this problem, but most existing compressors fail to preserve the topology of scientific data, leading to inaccuracies in downstream analyses and potentially erroneous scientific conclusions. In this work, we present a methodology for fully preserving the topology, specifically, Morse–Smale complexes (MSCs), in lossy-compressed 2D and 3D scalar field data from scientific simulations. We generalize the edit-based strategy introduced in MSz [1] (a previous method that preserves only segmentations and cannot preserve saddles or separatrices) by extending the framework to the full MSCs, including all critical points and separatrices. Our approach corrects the MSCs in the decompressed output of any error-bounded lossy compressor (e.g., SZ3 or ZFP), referred to as the base compressor, using an iterative editing strategy that preserves all critical points and their connectivity via separatrices. During compression, we generate a sequence of quantized edits that are applied to the decompressed output, ensuring accurate preservation of topological features while maintaining the error within prescribed bounds. The strategy iteratively fixes critical points and separatrices in alternating steps until convergence is achieved in a finite number of iterations. To meet diverse application needs, our method offers flexible options (e.g., whether to preserve the geometry of separatrices) that balance compression efficiency with feature preservation. To reduce computation time, we leverage GPU parallelism to accelerate each component of the workflow. Experiments on multiple datasets demonstrate that our method achieves 100% preservation of Morse–Smale complexes.

Abstract:
Redirected walking (RDW) enables users to walk naturally in a virtual environment (VE) that is larger than the physical environment (PE). However, relying entirely on redirected walking to navigate large VEs can be challenging. Teleportation, on the other hand, is a popular locomotion technique in VR as it enables rapid travel over long distances without inducing VR sickness. It is thus a practical approach to combine RDW and teleportation for VR navigation, using RDW for nearby destinations and teleportation for distant targets. This study introduces Teleportation with Reorientation (TRO), a technique that inlays reorientation into teleportation to enhance the performance of redirected walking. TRO applies an orientation offset immediately after teleportation, which users counteract by physically turning in place, thereby creating an opportunity to realign their heading to a favorable direction in the PE. A simulation-based evaluation and a user study were conducted to compare TRO against regular teleportations, all used in conjunction with RDW. The results demonstrated that TRO significantly improved the performance of redirected walking compared to regular teleportation. Regarding the overall navigation experience, even though TRO resulted in a higher workload and lower usability because it uses a perceptible reorientation strategy, most participants still considered TRO to be acceptable. In addition, participants indicated that their likeability and preference ratings for the TRO variant featuring one-step turning were the same as those for regular teleportation. These findings suggest that TRO with one-step turning can function as a viable alternative to regular teleportation for VR navigation when both RDW and teleportation are available.

Abstract:
Reconstructing large-scale 3D scenes remains challenging due to the need to balance photorealistic quality, real-time rendering, and compact storage. Recent progress in 3D Gaussian Splatting (3DGS) has achieved impressive fidelity and speed, yet its large-scale application suffers from excessive primitive counts, leading to prohibitive storage and rendering costs. To overcome this inefficiency, we introduce a novel semantic-guided hybrid representation that unifies textured meshes and 3D Gaussians in a differentiable framework. The key idea is to leverage meshes for geometrically regular regions such as roads and building facades, while reserving Gaussians for fine, complex details like vegetation. Our method is realized through three key technical contributions. First, we develop a semantic-guided adaptive modeling pipeline that fuses multi-view segmentation onto the scene mesh to robustly partition the scene and prune redundant Gaussians. Second, we introduce a high-performance CUDA-based hybrid renderer that seamlessly combines mesh rasterization with Gaussian splatting, enabling correct occlusion handling and joint optimization of both representations. Finally, we propose a mesh-guided sampling strategy that adaptively adds Gaussians to recover fine details in under-reconstructed areas. Extensive experiments on diverse large-scale datasets demonstrate that our approach significantly reduces storage requirements and accelerates rendering performance while maintaining comparable or superior visual quality.

Abstract:
SurfaceNets is a powerful visualization technique typically used to contour non-continuous, discrete, volumetric scalar fields such as segmentation label maps. Label maps are ubiquitous to medical computing, biological studies, and materials characterization, used in applications ranging from anatomical atlas creation to nanotechnology analysis. Due to the uniform spacing of volume data, however, representing data with highly variable resolution is challenging. Consequently we have developed a generalized high-performance, parallel SurfaceNets algorithm that processes unorganized, labeled point clouds. Based on a scalable, meshless Voronoi approach, the algorithm independently processes each Voronoi hull in parallel using a hierarchical neighborhood point search metric. By employing novel topological constructs, the resulting meshless tessellation can be readily transformed into a connected conformal mesh, from which multiple, valid contour surfaces can be simultaneously extracted and smoothed. Additional contributions include a general API for locating points proximal to Voronoi hulls; the definition of topological coordinates used to detect and eliminate numerical degeneracies, merge coincident points, rapidly produce the dual Delaunay triangulation, and build smoothing stencils; and the construction of a Voronoi adjacency graph along with associated necessary conditions to ensure the generation of valid tessellations. Characterization of parallel performance is also quantified, including producing Voronoi and Delaunay tessellations of 128 million hulls and more than 750 million tetrahedra. A software implementation is available from the open source the Visualization Toolkit (VTK) system at vtk.org.

Abstract:
Mesh deformation is a fundamental tool in 3D content manipulation. Despite extensive prior research, existing approaches often suffer from low output quality, require significant manual tuning, or depend on data-intensive training. To address these limitations, we introduce a training-free, handle-based mesh deformation method. Our core idea is to leverage a Vision-Language Model (VLM) to interpret and manipulate a handle-based interface. We begin by applying cone singularity detection to identify a sparse set of potential handles. The VLM is then prompted to select both the deformable sub-parts of the mesh and the handles that best align with user instructions. Subsequently, we query the desired deformed positions of the selected handles in screen space. To reduce uncertainty inherent in VLM predictions, we aggregate the results from multiple camera views using a novel multi-view voting scheme. Across a suite of benchmarks, our method produces deformations that align more closely with user intent, as measured by CLIP and GPTEval3D scores, while introducing low distortion—quantified via membrane energy. In summary, our approach is highly automated and consistently delivers high-quality mesh deformations.

Abstract:
Existing 4D Gaussian Splatting methods typically rely on per-Gaussian deformation from a canonical space to target frames, which overlooks the strong redundancy among spatially and temporally adjacent Gaussian primitives and leads to suboptimal efficiency. To address this limitation, we propose ADC-GS++, an anchor-driven compact Gaussian splatting framework for efficient and high-quality dynamic scene reconstruction. Specifically, ADC-GS++ organizes Gaussian primitives into an anchor-based canonical representation, enabling attribute sharing across local regions. To efficiently model dynamic scenes, we introduce a static–dynamic decomposition mechanism and further employ a coarse-to-fine deformation strategy driven by dynamic anchors at multiple granularities. In addition, a unified rate-distortion optimization is adopted to achieve a balanced trade-off between storage efficiency and reconstruction fidelity. Furthermore, a temporal significance-based anchor refinement strategy is employed to dynamically grow and prune anchors, allowing robust adaptation to complex and large-scale motions. Extensive experiments on multiple real-world dynamic scene datasets demonstrate that ADC-GS++ significantly improves rendering speed over deformation-based approaches by 300%–700%, while maintaining competitive rendering quality. Moreover, ADC-GS++ achieves a more favorable rate-distortion trade-off, resulting in substantially reduced storage consumption across different bitrate settings.

Abstract:
Hairpin vortices are fundamental structures within turbulent boundary layers, playing a crucial role in energy dissipation, mixing, and momentum transport. However, accurately extracting these structures remains challenging due to their irregular shapes, varying scales, and entanglement with surrounding vortical structures. This article presents a novel framework for the extraction of hairpin vortices from turbulent boundary layers. The method begins by identifying vortical regions and decomposing them into smaller segments using merge tree–based segmentation. A novel bottom-up rejoining approach is then introduced to group candidate segments according to the geometric and physical characteristics of hairpin vortices, resulting in regions that encompass complete hairpin vortex structures. These regions are subsequently refined and validated through skeleton analysis to detect the characteristic hairpin shape and are further confirmed using additional scalar-based criteria. Finally, smooth enclosing surfaces are generated for effective visualization. To enable quantitative evaluation, reference hairpin vortices are extracted from several flow datasets and used as ground truth. Compared with existing approaches, the proposed method eliminates manual parameter tuning, reduces under- and over-segmentation, and significantly improves both accuracy and computational efficiency. Demonstrations on multiple turbulent flow cases show that the method is robust and effective for hairpin vortex extraction under varying boundary layer conditions.

Abstract:
Neural Radiance Field (NeRF) is a groundbreaking paradigm in neural implicit representations that revolutionized 3D reconstruction, rendering, and dynamic scene modeling. To address cross-domain fragmentation and unclear technical pathways, we present a systematic framework that surveys theoretical foundations, benchmark datasets, methodological advances, and application scenarios. We begin by analyzing NeRF’s core mechanisms, including radiance field modeling and differentiable volume rendering, and by defining standardized evaluation benchmarks. Then we chart evolutionary pathways in model optimization, input adaptation, and dynamic scene modeling and analyze how key methods are linked. Furthermore, we provide task-specific insights that highlight migration bottlenecks and potential remedies across digital content creation, embodied perception, and other specialized domains. We also provide comprehensive references and forward-looking guidance for further theoretical refinements and cross-disciplinary deployment of NeRF-based technologies.

Abstract:
Line art is an illustrative technique with a wide use in education and art. In the context of image abstraction, its potential for increasing memorisation and recognition has been demonstrated, which motivates its use in scientific illustrations. While much work has focused on the conversion of 3D models into a line-art representation, there is a lack of solutions for virtual reality. Applying existing methods for each eye independently turns out to fall short due to cost constraints, distracting artifacts due to inconsistencies, or limitations regarding the input geometry. To address these limitations, we present a contour renderer for virtual reality. It operates in screen space, making it flexible, yet it relies on a local surface approximation combined with a registration error metric for robustness. Inconsistent occluding contours are continuously merged, and lines with no correspondence between both eyes are culled. The method is easy to implement, highly efficient even for high-resolution imagery, and, according to user evaluations, avoids the noticeable artifacts produced by existing work.

Abstract:
With recent advances in information and communication technologies, Hybrid meetings, where local attendees are physically present and remote participants join virtually, have become increasingly common. However, remote participants often experience reduced contextual awareness and a sense of isolation. To address these issues, we propose HybridSphere, a hybrid meeting system that reconstructs a shared virtual reality environment for remote participants. The system employs a 360-degree camera and pose estimation to generate real-time avatar representations of local attendees to allow remote users with head-mounted displays to experience the meeting as if all participants are in the same virtual space. We conducted a user study comparing HybridSphere to a baseline condition in which remote participants viewed an unmodified 360-degree video. Although the avatars were rated lower in perceived trustworthiness and likability due to limited visual fidelity, participants appreciated the seated-6DoF function. These results suggest that immersive and spatially flexible VR representations can enhance remote engagement in hybrid meetings.

Abstract:
World-in-Miniatures (WIMs) are an established technique in virtual reality (VR) that provide an overview of the surrounding virtual environment. While traditionally used to support navigation over larger distances and distant manipulation in large-scale spaces, their potential for comparing alternative versions of a virtual environment has not been exploited. In this paper, we introduce a Multi-WIM paradigm that facilitates the detailed exploration and comparison of urban planning alternatives. Each alternative is visualized in a separate WIM. The WIMs can be arranged side by side (juxtaposed) or stacked. Stacking allows toggling between alternatives or using a swipe tool where on each side one alternative is shown. These three modes (juxtaposed, toggle, swipe) provide flexible ways of exploring differences and similarities between alternatives, with highlighting functionalities that reveal differences of functional units across designs. We evaluated the Multi-WIM paradigm in a single-factor VR user study (N=24), combining a single-user experiment with a multi-user exploration. Participants had to identify differences and estimate distances, areas and heights across urban planning alternatives. Results show that the stacked toggle mode was significantly faster for finding differences while the juxtaposed view better supported a landmark recall task. Furthermore, toggle was the overall preferred technique in the single-user mode. In contrast, in multi-user settings participants favored juxtaposed, as it fostered collaborative discussion and shared sensemaking. The identified design trade-offs between efficiency and collaboration indicate the need to incorporate all three WIM modes, together with seamless transitions among them, in urban planning applications so as to accommodate both loosely and tightly coupled phases of collaboration.

Abstract:
Virtual reality (VR) enables interactions beyond real-world limitations, yet many such capabilities remain underexplored due to concerns about user comfort and increased disorientation and cybersickness. This study examined how altering the user's virtual pitch orientation influences interaction performance and subjective ratings. In a within-subject design, 30 seated participants experienced 12 virtual tilts from moderate to extreme (±180°). We assessed comfort, simulator sickness, perceptual responses, and performance. Our results revealed no significant increases in nausea, disorientation, or decreased comfort, with mild tilts performing equivalently to baseline. Notably, even extreme tilts maintained low nausea levels. Performance effects were mixed: forward tilts yielded similar or slightly improved performance, whereas backward tilts modestly impaired performance, though not significantly. These findings suggest that VR experiences featuring virtual body orientations distinct from the user's physical posture are feasible without compromising comfort or performance, enabling novel opportunities in simulation, interface design, visualization, creative content creation, and VR gaming.

Abstract:
Neural shape representation generally refers to representing 3D geometry using neural networks, e.g., computing a signed distance or occupancy value at a specific spatial position. In this paper we present a neural-network architecture suitable for accurate encoding of 3D shapes in a single forward pass. Our architecture is based on a multi-scale hybrid system incorporating graph-based and voxel-based components, as well as a continuously differentiable decoder. The hybrid system includes a novel way of voxelizing point-based features in neural networks by projecting the point “feature-field” onto a grid. This projection is insensitive to local point density, and we show that it can be used to obtain smoother and more detailed reconstructions, in particular when combined with oriented point clouds as input. Our architecture also requires only a single forward pass, instead of the latent-code optimization used in auto-decoder methods. Furthermore, our network is trained to solve the well-established eikonal equation and only requires knowledge of the zero-level set for training and inference. We additionally propose a modification to the aforementioned loss function for the case that surface normals are not well defined, e.g., in the context of non-watertight surfaces and non-manifold geometry. Overall, our method consistently outperforms other baselines on the surface reconstruction task across a wide variety of datasets, while being more computationally efficient and requiring fewer parameters.

Abstract:
Cybersickness, or sickness caused by virtual reality (VR), represents a significant threat to the usability of VR applications. Repeated exposure to the same VR stimulus causes a reduction in cybersickness, referred to as Cybersickness Abatement from Repeated Exposure (CARE). This study examined whether the benefits of CARE generalize across distinct VR contexts, which was operationalized as three distinct games (a climbing game, a puzzle game, and a stealth survival game). Participants played a VR game for up to 20 minutes. Those in the Repeated Exposure condition played one VR game (either a puzzle game or a climbing game) on three separate days followed by a different VR game (a survival game) on the fourth day. Those in the Single Exposure condition played the survival game once. The three games all differed in several ways, including environment and task, whereas the puzzle and survival games shared a similar joystick locomotion interface that differed from the locomotion interface in the climbing game. Results indicate that cybersickness on Day 4 of the Repeated Exposure condition was significantly lower than that in the Single Exposure condition, regardless of which game was experienced on Days 1-3. The practical implication of this finding is that CARE that occurs in one VR context can generalize to a novel context with a distinct environment, task, and locomotion interface. Results are considered in the context of multiple theoretical explanations for CARE, including sensory rearrangement and habituation. These results support systematic exposure as an approach to reducing cybersickness.

Abstract:
Empirical research on perception and cognition has laid the foundation for visualization design, often distilled into practical guidelines intended to support effective chart creation. However, it remains unclear how well these research-driven insights are reflected in the guidelines practitioners actually use. In this paper, we investigate the research-practice gap in visualization design guidelines through a mixed-methods approach. We first collected design guidelines from practitioner-facing sources and empirical studies from academic venues to assess their alignment. To complement this analysis, we conducted surveys and interviews with practitioners and researchers to examine their experiences, perceptions, and challenges surrounding the development and use of design guidelines. Our findings reveal misalignment between empirical evidence and widely used guidelines, differing perspectives between communities, and key barriers that contribute to the persistence of the research-practice gap.

Abstract:
The growing demand for diverse and realistic character animations in video games and films has driven the development of natural language-controlled motion generation systems. While recent advances in text-driven 3D human motion synthesis have made significant progress, generating realistic multi-person interactions remains a major challenge. Existing methods, such as denoising diffusion models and autoregressive frameworks, have explored interaction dynamics using attention mechanisms and causal modeling. However, they consistently overlook a critical physical constraint: the explicit spatial distance between interacting body parts, which is essential for producing semantically accurate and physically plausible interactions. To address this limitation, we propose InterDist, a novel masked generative Transformer model operating in a discrete state space. Our key idea is to decompose two-person motion into three components: two independent, interaction-agnostic single-person motion sequences and a separate interaction distance sequence. This formulation enables direct learning of both individual motion and dynamic spatial relationships from text prompts. We implement this via a VQ-VAE that jointly encodes independent motions and relative distances into discrete codebooks, followed by a bidirectional masked generative Transformer that models their joint distribution conditioned on text. To better align motion and language, we also introduce a cross-modal interaction module to enhance text-motion association. Our approach ensures the generated motions exhibit both semantic alignment with textual descriptions and preserving plausible inter-character distances, setting a new benchmark for text-driven multi-person interaction generation.

Abstract:
3D Gaussian Splatting (3DGS) has shown great promise in a variety of applications due to its exceptional real-time rendering quality and explicit representation, leading to numerous improvements across various fields. However, existing methods lack consideration of main objects and important structural information in their overall optimization strategies. This results in blurring of main objects in adaptive rendering and the loss of high-frequency details on targets that are insufficiently captured. In this work, we introduce a semantic-guided 3DGS method with adaptive rendering, which optimizes important structures through the guidance of boundary Gaussians, while leveraging semantic features to enhance the rendering of main objects in adaptive rendering. Experiments show that the proposed semantic-guided method can enhance important structures and high-frequency information in corner regions without significantly increasing the total number of Gaussians. This method also improves the separability between objects. At the same time, a semantic-guided Level-of-Detail (LoD) rendering approach enables the rapid display of main targets and the rendering of a complete scene. The semantic-guided methodology we have presented exhibits compatibility with a range of existing techniques.

Abstract:
We present an extension to traditional rig skinning, like Linear Blend Skinning (LBS), to produce secondary motions that exhibit the appearance of a physical phenomena without need for simulation. At the core of the technique, we call dynamic skinning, is a set of deformers which offset position of individual vertices as a function of position derivatives and time. Examples of such deformers create effects such as oscillation in response to movement and the appearance of wave propagation, among others. Because the technique computes offsets directly and does not solve physics equations, it is extremely fast to compute. It also boasts a highdegree of customizability which supports a desirable artist workflow and fine level of control. Finally, we showcase the technique in a number of scenarios and make comparisons with the state of the art.

Abstract:
How audiences read, interpret, and critique data visualizations is mainly assessed through performance tests featuring tasks like value retrieval. Yet, other factors shown to shape visualization understanding, such as numeracy, graph familiarity, and aesthetic perception, remain underrepresented in existing instruments. To address this, we design and test a Multidimensional Assessment Method for Visualization Understanding (MdamV). This method integrates task-based measures with self-perceived ability ratings and open-ended critique, applied directly to the visualizations being read. Grounded in learning sciences frameworks that view understanding as a multifaceted process, MdamV spans six dimensions: Comprehending, Decoding, Aestheticizing, Critiquing, Reading, and Contextualizing. Validation was supported by a survey (N = 438) representative of Austria’s population (ages 18–74, male/female split), using a line chart and a bar chart on climate data. Findings show, for example, that about a quarter of respondents indicate deficits in comprehending simple data units, roughly one in five people felt unfamiliar with each chart type, and self-assessed numeracy was significantly related to data reading performance (p = 0.0004). Overall, the evaluation of MdamV demonstrates the value of assessing visualization understanding beyond performance, framing it as a situated process tied to particular visualizations

Abstract:
To address the issue of style expression in existing text-driven human motion synthesis methods, we propose DiFusion, a framework for diversely stylized motion generation. It offers flexible control of content through texts and style via multiple modalities, i.e., textual labels or motion sequences. Our approach employs a dual-condition motion latent diffusion model, enabling independent control of content and style through flexible input modalities. To tackle the issue of imbalanced complexity between the text-motion and style-motion datasets, we propose the Digest-and-Fusion training scheme, which digests domain-specific knowledge from both datasets and then adaptively fuses them into a compatible manner. Comprehensive evaluations demonstrate the effectiveness of our method and its superiority over existing approaches in terms of content alignment, style expressiveness, realism, and diversity. Additionally, our approach can be extended to practical applications, such as motion style interpolation.

Abstract:
This paper presents a novel topology-aware dimensionality reduction approach aiming at accurately visualizing the cyclic patterns present in high dimensional data. To that end, we build on the Topological Autoencoders (TopoAE) (Moor et al., 2020) formulation. First, we provide a novel theoretical analysis of its associated loss and show that a zero loss indeed induces identical persistence pairs (in high and low dimensions) for the 0-dimensional persistent homology (\textPH^0) of the Rips filtration. We also provide a counter example showing that this property no longer holds for a naive extension of TopoAE to \textPH^d for d\geq 1. Based on this observation, we introduce a novel generalization of TopoAE to 1-dimensional persistent homology (\textPH^1), called TopoAE++, for the accurate generation of cycle-aware planar embeddings, addressing the above failure case. This generalization is based on the notion of cascade distortion, a new penalty term favoring an isometric embedding of the 2-chains filling persistent 1-cycles, hence resulting in more faithful geometrical reconstructions of the 1-cycles in the plane. We further introduce a novel, fast algorithm for the exact computation of \textPH^ for Rips filtrations in the plane, yielding improved runtimes over previously documented topology-aware methods. Our method also achieves a better balance between the topological accuracy, as measured by the Wasserstein distance, and the visual preservation of the cycles in low dimensions.

Abstract:
Generalizable NeRF synthesizes novel views of unseen scenes without per-scene training. The view-epipolar transformer has become popular in this field for its ability to produce high-quality views. Existing methods with this architecture rely on the assumption that texture consistency across views can identify object surfaces, with such identification crucial for determining where to reconstruct texture. However, this assumption is not always valid, as different surface positions may share similar texture features, creating ambiguity in surface identification. To handle this ambiguity, this paper introduces 3D volume features into the view-epipolar transformer. These features contain geometric information, which will be a supplement to texture features. By incorporating both texture and geometric cues in consistency measurement, our method mitigates the ambiguity in surface detection. This leads to more accurate surfaces and thus better novel view synthesis. Additionally, we propose a decoupled decoder where volume and texture features are used for density and color prediction respectively. In this way, the two properties can be better predicted without mutual interference. Experiments show improved results over existing transformer-based methods on both real-world and synthetic datasets.

Abstract:
Interactive 3D model texture editing presents enhanced opportunities for creating 3D assets, with freehand drawing style offering the most intuitive experience. However, existing methods primarily support sketch-based interactions for outlining, while the utilization of coarse-grained scribble-based interaction remains limited. Furthermore, current methodologies often encounter challenges due to the abstract nature of scribble instructions, which can result in ambiguous editing intentions and unclear target semantic locations. To address these issues, we propose ScribbleSense, an editing method that combines multimodal large language models (MLLMs) and image generation models to effectively resolve these challenges. We leverage the visual capabilities of MLLMs to predict the editing intent behind the scribbles. Once the semantic intent of the scribble is discerned, we employ globally generated images to extract local texture details, thereby anchoring local semantics and alleviating ambiguities concerning the target semantic locations. Experimental results indicate that our method effectively leverages the strengths of MLLMs, achieving state-of-the-art interactive editing performance for scribble-based texture editing.

Abstract:
Understanding and distinguishing temporal patterns in time series data is essential for scientific discovery and decision-making. For example, in biomedical research, uncovering meaningful patterns in physiological signals can improve diagnosis, risk assessment, and patient outcomes. However, existing methods for time series pattern discovery face major challenges, including high computational complexity, limited interpretability, and difficulty in capturing meaningful temporal structures. To address these gaps, we introduce a novel learning framework that jointly trains two Transformer models using complementary time series representations: shapelet-based representations to capture localized temporal structures and traditional feature engineering to encode statistical properties. The learned shapelets serve as interpretable signatures that differentiate time series across classification labels. Additionally, we develop a visual analytics system—SigTime—with coordinated views to facilitate exploration of time series signatures from multiple perspectives, aiding in useful insights generation. We quantitatively evaluate our learning framework on eight publicly available datasets and one proprietary clinical dataset. Additionally, we demonstrate the effectiveness of our system through two usage scenarios along with the domain experts: one involving public ECG data and the other focused on preterm labor analysis.

Abstract:
Correlation clustering (CC) offers an effective approach to analyze scalar field ensembles by detecting correlated regions and consistent structures, enabling the extraction of meaningful patterns. However, existing CC methods are computationally expensive, making them impractical for both interactive analysis and large-scale scalar fields. We introduce the Local-to-Global Correlation Clustering (LoGCC) framework, which accelerates pivot-based CC by leveraging the spatial structure of scalar fields and the weak transitivity of correlation. LoGCC operates in two stages: a local step that uses the neighborhood graph of the scalar field's spatial domain to build highly correlated local clusters, and a global step that merges them into global clusters. We implement the LoGCC framework for two well-known pivot-based CC methods, Pivot and CN-Pivot, demonstrating its generality. Our evaluation using synthetic and real-world meteorological and medical image segmentation datasets shows that LoGCC achieves speedups—up to 15 × for Pivot and 200 × for CN-Pivot—and improved scalability to larger scalar fields, while maintaining cluster quality. These contributions broaden the applicability of correlation clustering in large-scale and interactive analysis settings.

Abstract:
Contemporary discourse on data communication has discussed logos (reason) and, more recently, ethos (credibility) extensively. While the concept of pathos (emotional appeal) has received growing attention in the visualization community in recent years, its connection to related concepts such as rhetoric and aesthetics remains underexplored. In this paper, we provide working definitions of these terms, contextualize them within data visualization, and explore their overlaps and differences in light of their historical development. This historical perspective offers a more holistic understanding of how these approaches in science and philosophy have evolved over time, contributing to a deeper comprehension of their integration into the design process. Drawing on Campbell's seven circumstances, we illustrate how pathos functions as a rhetorical strategy in contemporary data visualizations, examining the interplay of rhetorical strategies, aesthetic qualities, and offering our interpretation of how these elements work together.

Abstract:
Comparing tensors and identifying their (dis)similar structures is fundamental in understanding the underlying phenomena for complex data. Tensor decomposition methods help analysts extract tensors' essential characteristics and aid in visual analytics for tensors. In contrast to dimensionality reduction (DR) methods designed only for analyzing a matrix (i.e., second-order tensor), existing tensor decomposition methods do not support flexible comparative analysis. To address this analysis limitation, we introduce a new tensor decomposition method, named tensor unified linear comparative analysis (TULCA), by extending its DR counterpart, ULCA, for tensor analysis. TULCA integrates discriminant analysis and contrastive learning schemes for tensor decomposition, enabling flexible comparison of tensors. We also introduce an effective method to visualize a core tensor extracted from TULCA into a set of 2D visualizations. We integrate TULCA's functionalities into a visual analytics interface to support analysts in interpreting and refining the TULCA results. We demonstrate the efficacy of TULCA and the visual analytics interface with computational evaluations and two case studies, including an analysis of log data collected from a supercomputer.

Abstract:
Overdraw is inevitable in large-scale scatterplots. Current scatterplot abstraction methods lose features in medium-to-low density regions. We propose a visual abstraction method designed to provide better feature preservation across arbitrary abstraction levels for large-scale scatterplots, particularly in medium-to-low density regions. The method consists of three closely interconnected steps: first, we partition the scatterplot into iso-density regions and equalize visual density; then, we allocate pixels for different classes within each region; finally, we reconstruct the data distribution based on pixels. User studies, quantitative and qualitative evaluations demonstrate that, compared to previous methods, our approach better preserves features and exhibits a special advantage when handling ultra-high dynamic range data distributions.

Abstract:
Visualizing multiple time series presents fundamental tradeoffs between scalability and visual clarity. Time series capture the behavior of many large-scale real-world processes, from stock market trends to urban activities. Users often gain insights by visualizing them as line charts, juxtaposing or superposing multiple time series to compare them and identify trends and patterns. However, existing representations struggle with scalability: when covering long time spans, leading to visual clutter from too many small multiples or overlapping lines. We propose TiVy, a new algorithm that summarizes time series using sequential patterns. It transforms the series into a set of symbolic sequences based on subsequence visual similarity using Dynamic Time Warping (DTW), then constructs a disjoint grouping of similar subsequences based on the frequent sequential patterns. The grouping result, a visual summary of time series, provides uncluttered superposition with fewer small multiples. Unlike common clustering techniques, TiVy extracts similar subsequences (of varying lengths) aligned in time. We also present an interactive time series visualization that renders large-scale time series in real-time. Our experimental evaluation shows that our algorithm (1) extracts clear and accurate patterns when visualizing time series data, (2) achieves a significant speed-up (1000×) compared to a straightforward DTW clustering. We also demonstrate the efficiency of our approach to explore hidden structures in massive time series data in two usage scenarios.

Abstract:
Text-to-3D (T23D) generation has transformed digital content creation, yet remains bottlenecked by blind trial-and-error prompting processes that yield unpredictable results. While visual prompt engineering has advanced in text-to-image domains, its application to 3D generation presents unique challenges requiring multi-view consistency evaluation and spatial understanding. We present Sel3DCraft, a visual prompt engineering system for T23D that transforms unstructured exploration into a guided visual process. Our approach introduces three key innovations: a dual-branch structure combining retrieval and generation for diverse candidate exploration; a multi-view hybrid scoring approach that leverages MLLMs with innovative high-level metrics to assess 3D models with human-expert consistency; and a prompt-driven visual analytics suite that enables intuitive defect identification and refinement. Extensive testing and a user study demonstrate that Sel3DCraft surpasses other T23D systems in supporting creativity for designers.

Abstract:
Text labels are widely used to convey auxiliary information in visualization and graphic design. The substantial variability in the categories and structures of labeled objects leads to diverse label layouts. Recent single-model learning-based solutions in label placement struggle to capture fine-grained differences between these layouts, which in turn limits their performance. In addition, although human designers often consult previous works to gain design insights, existing label layouts typically serve merely as training data, limiting the extent to which embedded design knowledge can be exploited. To address these challenges, we propose a mixture of cluster-guided experts (MoCE) solution for label placement. In this design, multiple experts jointly refine layout features, with each expert responsible for a specific cluster of layouts. A cluster-based gating function assigns input samples to experts based on representation clustering. We implement this idea through the Label Placement Cluster-guided Experts (LPCE) model, in which a MoCE layer integrates multiple feed-forward networks (FFNs), with each expert composed of a pair of FFNs. Furthermore, we introduce a retrieval augmentation strategy into LPCE, which retrieves and encodes reference layouts for each input sample to enrich its representations. Extensive experiments demonstrate that LPCE achieves superior performance in label placement, both quantitatively and qualitatively, surpassing a range of state-of-the-art baselines. Our algorithm is available at https://github.com/PingshunZhang/LPCE.

Abstract:
Visual analytics (VA) is typically applied to complex data, thus requiring complex tools. While visual analytics empowers analysts in data analysis, analysts may get lost in the complexity occasionally. This highlights the need for intelligent assistance mechanisms. However, even the latest LLM-assisted VA systems only provide help when explicitly requested by the user, making them insufficiently intelligent to offer suggestions when analysts need them the most. We propose a ProactiveVA framework in which LLM-powered UI agent monitors user interactions and delivers context-aware assistance proactively. To design effective proactive assistance, we first conducted a formative study analyzing help-seeking behaviors in user interaction logs, identifying when users need proactive help, what assistance they require, and how the agent should intervene. Based on this analysis, we distilled key design requirements in terms of intent recognition, solution generation, interpretability and controllability. Guided by these requirements, we develop a three-stage UI agent pipeline including perception, reasoning, and acting. The agent autonomously perceives users' needs from VA interaction logs, providing tailored suggestions and intuitive guidance through interactive exploration of the system. We implemented the framework in two representative types of VA systems, demonstrating its generalizability, and evaluated the effectiveness through an algorithm evaluation, case and expert study and a user study. We also discuss current design trade-offs of proactive VA and areas for further exploration.

Abstract:
Dimensionality reduction techniques help analysts make sense of complex, high-dimensional spatial datasets, such as multiplexed tissue imaging, satellite imagery, and astronomical observations, by projecting data attributes into a two-dimensional space. However, these techniques typically abstract away crucial spatial, positional, and morphological contexts, complicating interpretation and limiting insights. To address these limitations, we present SEAL, an interactive visual analytics system designed to bridge the gap between abstract 2D embeddings and their rich spatial imaging context. SEAL introduces a novel hybrid-embedding visualization that preserves image and morphological information while integrating critical high-dimensional feature data. By adapting set visualization methods, SEAL allows analysts to identify, visualize, and compare selections-defined manually or algorithmically-in both the embedding and original spatial views, facilitating a deeper understanding of the spatial arrangement and morphological characteristics of entities of interest. To elucidate differences between selected sets of items, SEAL employs a scalable surrogate model to calculate feature importance scores, identifying the most influential features governing the position of objects within embeddings. These importance scores are visually summarized across selections, with mathematical set operations enabling detailed comparative analyses. We demonstrate SEAL's effectiveness and versatility through three case studies: colorectal cancer tissue analysis with a pharmacologist, melanoma investigation with a cell biologist, and exploration of sky survey data with an astronomer. These studies underscore the importance of integrating image context into embedding spaces when interpreting complex imaging datasets. Implemented as a standalone tool while also integrating seamlessly with computational notebooks, SEAL provides an interactive platform for spatially informed exploration of high-dimensional datasets, significantly enhancing interpretability and insight generation.

Abstract:
An essential task in analyzing collaborative design processes, such as those that are part of workshops in design studies, is identifying design outcomes and understanding how the collaboration between participants formed the results and led to decision-making. However, findings are typically restricted to a consolidated textual form based on notes from interviews or observations. A challenge arises from integrating different sources of observations, leading to large amounts and heterogeneity of collected data. To address this challenge we propose a practical, modular, and adaptable framework of workshop setup, multimodal data acquisition, AI-based artifact extraction, and visual analysis. Our interactive visual analysis system, reCAPit, allows the flexible combination of different modalities, including video, audio, notes, or gaze, to analyze and communicate important workshop findings. A multimodal streamgraph displays activity and attention in the working area, temporally aligned topic cards summarize participants' discussions, and drill-down techniques allow inspecting raw data of included sources. As part of our research, we conducted six workshops across different themes ranging from social science research on urban planning to a design study on band-practice visualization. The latter two are examined in detail and described as case studies. Further, we present considerations for planning workshops and challenges that we derive from our own experience and the interviews we conducted with workshop experts. Our research extends existing methodology of collaborative design workshops by promoting data-rich acquisition of multimodal observations, combined AI-based extraction and interactive visual analysis, and transparent dissemination of results.

Abstract:
Text is an integral but understudied component of visualization design. Although recent studies have examined how text elements (e.g., titles and annotations) influence comprehension, preferences, and predictions, many questions remain about textual design and use in practice. This paper introduces a framework for understanding text functions in information visualizations, building on and filling gaps in prior classifications and taxonomies. Through an analysis of 120 real-world visualizations and 804 text elements, we identified ten distinct text functions, ranging from identifying data mappings to presenting valenced subtext. We further identify patterns in text usage and conduct a factor analysis, revealing four overarching text-informed design strategies: Attribution and Variables, Annotation-Centric Design, Visual Embellishments, and Narrative Framing. In addition to these factors, we explore features of title rhetoric and text multifunctionality, while also uncovering previously unexamined text functions, such as text replacing visual elements. Our findings highlight the flexibility of text, demonstrating how different text elements in a given design can combine to communicate, synthesize, and frame visual information. This framework adds important nuance and detail to existing frameworks that analyze the diverse roles of text in visualization.

Abstract:
Team-based combat scenarios are prevalent in various real-world applications like video gaming. Analyzing tactics in these scenarios is essential for gaining insights into game processes and improving combat behaviors. The decision-making data in team-based combat include character actions, movement trajectories, and event sequences. Existing studies face challenges in visualizing and analyzing combat tactics due to the complexity and the multifaceted characteristics of the decision-making data. To address these challenges, we introduce TactiVis, a visual analytics system designed for analyzing combat decision-making behavior. Using MOBA game as a representative case of team-based combat, TactiVis adopts a macro-to-micro tactics visual analytics framework consisting of three stages: match-level analysis, event-level understanding, and character-level comparison. In the TactiVis system, we introduce the v-storyline visualization, which encodes positions along the vertical axis to reveal tactical patterns. Case studies and a usability study demonstrate the utility and usability of TactiVis for helping analysts understand combat patterns and analyze tactics.

Abstract:
Hybrid rice breeding crossbreeds different rice lines and cultivates the resulting hybrids in fields to select those with desirable agronomic traits, such as higher yields. Recently, genomic selection has emerged as an efficient way for hybrid rice breeding. It predicts the traits of hybrids based on their genes, which helps exclude many undesired hybrids, largely reducing the workload of field cultivation. However, due to the limited accuracy of genomic prediction models, breeders still need to combine their experience with the models to identify regulatory genes that control traits and select hybrids, which remains a time-consuming process. To ease this process, in this paper, we proposed a visual analysis method to facilitate interactive hybrid rice breeding. Regulatory gene identification and hybrid selection naturally ensemble a dual-analysis task. Therefore, we developed a parametric dual projection method with theoretical guarantees to facilitate interactive dual analysis. Based on this dual projection method, we further developed a gene visualization and a hybrid visualization to verify the identified regulatory genes and hybrids. The effectiveness of our method is demonstrated through the quantitative evaluation of the parametric dual projection method, identified regulatory genes and desired hybrids in the case study, and positive feedback from breeders.

Abstract:
Constructing expressive and legible visualizations is a key activity for visualization designers. While numerous design guidelines exist, research on how specific graphical features affect perceived visual complexity remains limited. In this paper, we report on a crowdsourced study to collect human ratings of perceived complexity for diverse visualizations. Using these ratings as ground truth, we then evaluated three methods to estimate this perceived complexity: image analysis metrics, multilinear regression using manually coded visualization features, and automated feature extraction using a large language model (LLM). Image complexity metrics showed no correlation with human-perceived visualization complexity. Manual feature coding produced a reasonable predictive model but required substantial effort. In contrast, a zero-shot LLM (GPT-4o mini) demonstrated strong capabilities in both rating complexity and extracting relevant features. Our findings suggest that visualization complexity is truly in the eye of the beholder, yet can be effectively approximated using zero-shot LLM prompting, offering a scalable approach for evaluating the complexity of visualizations. The dataset and code for the study and data analysis can be found at https://osf.io/w85a4/.

Abstract:
In contemporary information ecologies saturated with misinformation, disinformation, and a distrust of science itself, public data communication faces significant hurdles. Although visualization research has broadened criteria for effective design, governing paradigms privilege the accurate and efficient transmission of data. Drawing on theory from linguistic anthropology, we argue that such approaches-focused on encoding and decoding propositional content-cannot fully account for how people engage with visualizations and why particular visualizations might invite adversarial or receptive responses. In this paper, we present evidence that data visualizations communicate not only semantic, propositional meaning-meaning about data-but also social, indexical meaning-meaning beyond data. From a series of ethnographically-informed interviews, we document how readers make rich and varied assessments of a visualization's “vibes”-inferences about the social provenance of a visualization based on its design features. Furthermore, these social attributions have the power to influence reception, as readers' decisions about how to engage with a visualization concern not only content, or even aesthetic appeal, but also their sense of alignment or disalignment with the entities they imagine to be involved in its production and circulation. We argue these inferences hinge on a function of human sign systems that has thus far been little studied in data visualization: socio-indexicality, whereby the formal features (rather than the content) of communication evoke social contexts, identities, and characteristics. Demonstrating the presence and significance of this socio-indexical function in visualization, this paper offers both a conceptual foundation and practical intervention for troubleshooting breakdowns in public data communication.

Abstract:
The dissemination of visualizations is primarily in the form of raster images, which often results in the loss of critical information such as source code, interactive features, and metadata. While previous methods have proposed embedding metadata into images to facilitate Visualization Image Data Retrieval (VIDR), most existing methods lack practicability since they are fragile to common image tampering during online distribution such as cropping and editing. To address this issue, we propose VisGuard, a tamper-resistant VIDR framework that reliably embeds metadata link into visualization images. The embedded data link remains recoverable even after substantial tampering upon images. We propose several techniques to enhance robustness, including repetitive data tiling, invertible information broadcasting, and an anchor-based scheme for crop localization. VisGuard enables various applications, including interactive chart reconstruction, tampering detection, and copyright protection. We conduct comprehensive experiments on VisGuard's superior performance in data retrieval accuracy, embedding capacity, and security against tampering and steganalysis, demonstrating VisGuard's competence in facilitating and safeguarding visualization dissemination and information conveyance.

Abstract:
Multiphase and multi-component fluid flows are highly intricate, exhibiting dynamic behaviors that challenge conventional simulation methods, especially across a wide range of operating conditions. Adding fluid-solid interactions further escalates this complexity, requiring simulations that are both physically accurate and computationally more tractable. Existing methods often fall short in aspects of either accuracy, efficiency, or stability, and sometimes may also consume excessive memory, rendering them unsuitable for many practical applications. In this paper, we introduce a novel, more practical solution combining a collision-enhanced lattice Boltzmann method with a finite-volume approach. This hybrid solver is designed to effectively simulate multiphase, multi-component flows via the phase-field modeling with a novel wetting boundary treatment in the finite-volume framework, while significantly reducing memory consumption. It enables simulations of multiple fluid phases under varied conditions, capturing rich interfacial behaviors. Additionally, it incorporates an immersed-boundary method to efficiently handle complex fluid-solid interactions. Validation tests confirm that our method produces physically consistent results and demonstrates accuracy across multiple evaluations. We provide comprehensive simulation results, along with comparisons of memory consumption and efficiency, highlighting the practical utility and superiority of our solver over conventional methods.

Abstract:
Recent years have witnessed significant advancements in text-guided style transfer, primarily attributed to innovations in diffusion models. These models excel in conditional guidance, utilizing text or images to direct the sampling process. Traditional style transfer focuses on low-level visual features, such as brushstroke textures and color distributions, and appears more like applying an artistic filter to an image. Artistic attribute transfer, however, transcends the limitations of traditional style transfer by achieving the transfer of visual concepts from color and brushstrokes to high level aesthetic attributes such as composition, pose, and key semantic elements, resulting in more natural outcomes. Therefore, we propose an innovative text-to-image artistic attribute transfer framework named ArtCrafter. Specifically, we introduce an attention-based style extraction module, meticulously engineered to capture the subtle artistic attribute elements within an image. This module features a multi-layer architecture that leverages the capabilities of perceiver attention mechanisms to integrate fine-grained information. Additionally, we present a novel text-image aligning augmentation component that adeptly balances control over both modalities, enabling the model to efficiently map image and text embeddings into a shared feature space. We achieve this through attention operations that enable smooth information flow between modalities. Lastly, we incorporate an explicit modulation that seamlessly blends multimodal enhanced embeddings with original embeddings through an embedding reframing design, empowering the model to generate diverse outputs. Extensive experiments demonstrate that ArtCrafter yields impressive results in visual stylization, exhibiting exceptional levels of artistic attribute intensity, controllability, and diversity.

Abstract:
Creating detailed 3D characters from a single image remains challenging due to the difficulty in separating semantic components during generation. Existing methods often produce entangled meshes with poor topology, hindering downstream applications like rigging and animation. We introduce SeparateGen, a novel framework that generates high-quality 3D characters by explicitly reconstructing them as distinct semantic components (e.g., body, clothing, hair, shoes) from a single, arbitrary-pose image. SeparateGen first leverages a multi-view diffusion model to generate consistent multi-view images in a canonical A-pose. Then, a novel component-aware reconstruction model, SC-LRM, conditioned on these multi-view images, adaptively decomposes and reconstructs each component with high fidelity. To train and evaluate SeparateGen, we contribute SC-Anime, the first large-scale dataset of 7,580 anime-style 3D characters with detailed component-level annotations. Extensive experiments demonstrate that SeparateGen significantly outperforms state-of-the-art methods in both reconstruction quality and multi-view consistency. Furthermore, our component-based approach effectively resolves mesh entanglement issues, enabling seamless rigging and asset reuse. SeparateGen thus represents a step towards generating high-quality, application-ready 3D characters from a single image. The SC-Anime dataset and our code will be publicly released.

Abstract:
Redirected walking (RDW) subtly adjusts the user’s visual perspective on head-mounted displays during natural walking to reduce forced resets, thus enlarging the size of the virtual environment that can be explored beyond that of the physical environment. Alignment-based RDW controllers aim to minimize spatial discrepancies by optimizing the alignment between the user’s physical and virtual environments. We introduce a novel alignment-based method that dynamically calculates mapping functions between physical and virtual geometries to enhance the algorithm’s awareness of the RDW environments. To achieve this, we first construct an abstract model defining a mapping function between physical and virtual geometries and establish feasibility constraints in differential form. We then concretize this mapping, optimize it, and develop a practical implementation for dynamic geometric mapping in RDW. Our approach distinguishes itself by determining dense spatial mappings around the user, rather than aligning environments according to limited metrics. Through extensive testing, our algorithm has proven to markedly decrease reset incidents in natural walking, surpassing existing RDW controllers. The introduction of dynamic geometric mapping provides a fresh perspective, contributing significant insights and advancing the field.

Abstract:
In the digital age, readers value quantitative journalism that is clear, concise, analytical, and humancentred. To understand complex topics, they often piece together scattered facts from multiple articles. Visual storytelling can transform fragmented information into clear, engaging narratives, yet its use with unstructured online articles remains largely unexplored. To fill this gap, we present Compendia, an automated system that analyzes online articles in response to a user’s query and generates a coherent data story tailored to the user’s informational needs. Compendia addresses key challenges of storytelling from unstructured text through two modules covering: Online Article Retrieval, which gathers relevant articles; Data Fact Extraction, which identifies, validates, and refines quantitative facts; Fact Organization, which clusters and merges related facts into coherent thematic groups; and Visual Storytelling, which transforms the organized facts into narratives with visualizations in an interactive scrollytelling interface. We evaluated Compendia through a quantitative analysis, confirming the accuracy in fact extraction and organization, and through two user studies with 16 participants, demonstrating its usability, effectiveness, and ability to produce engaging visual stories for open-ended queries.

Abstract:
We present FISN, a generalizable novel view synthesis algorithm that enables feedforward inference of Neural Radiance Fields (NeRF) or 3D Gaussian Splatting (3DGS) from reference images. Unlike existing work that either separately model the 3D feature space on each view or process multiview reference features by 3D-point-based view aggregation, FISN integrates multi-reference 3D cost volumes into a unified high-dimensional entity. Specifically, we reconceptualize the generalizable novel view synthesis task as a feedforward process of FInding Spatial Neighborhoods across this unified 4D feature space, comprising both view and spatial dimensions, and introduce View-Spatial Convolutions for direct 4D feature aggregation. This enhances the correlation among multiview neighboring points in a window-to-window manner and incorporates 3D spatial awareness. However, this approach poses two intertwined challenges: high computational expense for high-dimensional features and degraded rendering performance with low-resolution features. To address these challenges, FISN constructs a new efficient convolution paradigm, Decomposable View-Spatial Convolution, which includes a Spatial Cross Decomposition strategy as well as a Feature Compression and Upscaling module. This paradigm maintains multiview geometric consistency better than existing decomposition methods and achieves a balance between efficiency and fine-grained spatial features. Furthermore, by integrating Depth Refinement modules based on this paradigm, FISN further improves global depth understanding. Comprehensive evaluations on mainstream datasets and benchmarks demonstrate that FISN achieves state-of-the-art performance for both NeRF and 3DGS, and remains robust in challenging scenarios where existing 3DGS-based methods struggle, such as those with noisy poses or dense references. The code will be released soon.

Affiliations: Research Centre for Art and Culture Technology, The Hong Kong Polytechnic University, Hong Kong SAR, China; Beijing Engineering Research Center of Mixed Reality and Advanced Display, School of Optics and Photonics, Zhengzhou Research Institute, Beijing Institute of Technology, Beijing, China; State Key Laboratory of Chips and Systems for Advanced Light Field Display, School of Optics and Photonics, Beijing Institute of Technology, Beijing, China; PolyU NVIDIA Joint Research Centre, The Hong Kong Polytechnic University, Hong Kong SAR, China

Abstract:
Micro-gestures provide a natural, efficient, and privacy-preserving input modality; however, existing techniques often depend on environmental conditions, limiting their robustness and applicability in real-world settings. In this work, we present three portable prototypes– - finger-cot, finger-worn, and surface-based—that integrate compact pressure array sensors to support environment-independent micro-gesture interaction. We further propose a deep learning–based recognition model that accurately classifies 14 micro-gestures by analyzing temporal pressure patterns. Building upon these components, we introduce PortInput, a real-time interactive system that enables robust micro-gesture tracking and detection. We conducted two user studies with augmented reality (AR) head-mounted displays (HMDs). The first study evaluates input performance under both sitting and walking conditions, while the second compares PortInput with a commercial pressure-based ring device. The results show that PortInput improves usability and user experience, achieves comparable accuracy, and enables faster input with lower perceived workload. Overall, PortInput have potential to offer efficient, robust, and comfortable input across diverse application scenarios—ranging from AR/Virtual Reality (VR) headsets to smart homes and in-car systems—even in noisy or cluttered environments. This work provides a foundation for integrating pressure array sensors into ring-based or other portable devices, advancing always-available micro-gesture interaction for ubiquitous computing environments.

Abstract:
We present GSwap, a novel consistent and realistic video head-swapping system empowered by dynamic neural Gaussian portrait priors, which significantly advances the state of the art in face and head replacement. Unlike previous methods that rely primarily on 2D generative models or 3D Morphable Face Models (3DMM), our approach overcomes their inherent limitations, including poor 3D consistency, unnatural facial expressions, and restricted synthesis quality. Moreover, existing techniques struggle with full head-swapping tasks due to insufficient holistic head modeling and ineffective background blending, often resulting in visible artifacts and misalignments. To address these challenges, GSwap introduces an intrinsic 3D Gaussian feature field embedded within a full-body SMPL-X surface, effectively elevating 2D portrait videos into a dynamic neural Gaussian field. This innovation ensures high-fidelity, 3D-consistent portrait rendering while preserving natural head-torso relationships and seamless motion dynamics. To facilitate training, we adapt a pretrained 2D portrait generative model to the source head domain using only a few reference images, enabling efficient domain adaptation. Furthermore, we propose a neural re-rendering strategy that harmoniously integrates the synthesized foreground with the original background, eliminating blending artifacts and enhancing realism. Extensive experiments demonstrate that GSwap surpasses existing methods in multiple aspects, including visual quality, temporal coherence, identity preservation, and 3D consistency.

Abstract:
To direct attention and communicate context, visualization designers often employ augmentations such as annotations, animated transitions, and stylistic layouts. While graphical perception principles can helpfully prescribe effective low-level representations of data, they lack guidance for augmentations that service higher-level communication goals. To bridge this gap, we contribute a design space that frames designer intent as viewer-oriented cognitive behaviors, grounding communicative aims in actionable visualization design techniques, including both visual encodings and augmentations. This design space consists of common augmentation tactics (annotation, animation, stylized encodings, etc.) that implement design strategies (spotlighting, sequencing, association, etc.) to achieve higher-level design goals (observe, interpret, introspect), often simultaneously. We demonstrate the analytic and generative value of our design space with examples across varied designer objectives. We discuss how our contributions help align designer intent with reader takeaways, pave the way for readers to learn more effectively, and enable future (semi-)automated systems to support visualization designers in achieving their communication goals.

Abstract:
Design spaces serve as conceptual frameworks that enable systematic exploration of possibilities and constraints for particular design problems. Despite growing recognition of their importance in visualization research, the community faces two main challenges: characterizing what constitute a design space, given the lack of consensus on its definition, and determining how to construct these spaces in the absence of established methodologies. To address the challenges, we first conducted a literature review of visualization design space research, identifying three distinct research threads. Focusing on the thread that views design spaces as multi-dimensional frameworks, we refined our corpus to 49 papers and developed a unified conceptualization of design spaces. Building on this foundation, we proposed a systematic approach to design space construction, synthesized from an analysis of practices spanning five phases: exploration, data collection, creation, evaluation, and communication.

Abstract:
In the realm of virtual avatar creation, accurate relighting capabilities are key to enhancing realism and immersion. We propose a novel pipeline for building personalized and relightable avatars from a monocular video captured under unknown lighting. This minimal input poses challenges in material entanglement and novel-view inconsistency. To tackle these, we introduce a disentangled dynamic 3D Gaussian representation that models diverse material properties and supports photorealistic rendering and animation via a parametric face model. To resolve material ambiguity under uncontrolled lighting, we train a 2D diffusion-based model to predict canonical-lighting images and physically-based material maps from casually lit portraits. These predictions serve as supervisory signals to guide the 3D disentanglement process. Additionally, we incorporate a 3D prior to enhance novel-view consistency, improving geometry and appearance in unseen views. Experiments demonstrate that our approach significantly boosts reconstruction quality and relighting fidelity, offering a practical and cost-effective solution for creating high-quality personalized avatars.

Abstract:
Generating continuous and expressive human motion from textual descriptions is a critical challenge in applications such as gaming and filmmaking. Existing methods often struggle to maintain global coherence, realistic frame continuity, and smooth transitions. To address these limitations, we propose FCMD, a novel diffusion-based model for generating cohesive motion sequences from fine-grained textual descriptions. FCMD introduces three key innovations: (1) Fine-grained Text Fusion, which integrates detailed textual cues with transitional narratives to enhance semantic consistency; (2) History Motion Guidance, ensuring motion accuracy and consistency across consecutive frames; and (3) Smooth Stitching Sampling, which leverages preceding and current motion information to achieve seamless transitions. Additionally, FCMD employs a large language model (LLM) to refine motion datasets by extracting fine-grained textual descriptions. Extensive experiments demonstrate that FCMD outperforms state-of-the-art methods in generating coherent, natural, and highly controllable motion sequences.

Abstract:
Gesture interaction is standard in modern extended reality head-mounted displays, e.g., Microsoft HoloLens 2 and Apple Vision Pro. These gestures come from multiple sources, such as elicitation studies, the application designer, or an expert. However, it is unclear if the source of the gesture affects the memorability of gestures (i.e., level of recall). In a user study, 35 participants used the HoloLens 2 to test three gestures set from different sources: 1) expert-defined, whereby we asked four experts to define gestures; 2) user-defined, whereby each participant defined their gestures; and 3) elicitation-defined, whereby we used gestures proposed by previous work. Our results showed that expert-defined gestures (72.73% recall) were more likely to be remembered than elicitation-defined (59.87% recall) gestures. Further, user-defined gestures (88.75% recall) were the most memorable. These data suggest that the source of gestures has a substantial impact on their memorability.

Abstract:
Despite significant progress in point cloud upsampling, most existing methods rely heavily on supervised training with paired data, which are often difficult to acquire. Moreover, the inherent lack of explicit connectivity in point clouds makes it difficult to achieve both continuous and uniform densification while accurately recovering fine geometric structures. To address these problems, we propose an upsampling model that treats this task as saliency-aware implicit surface sampling, enabling self-supervised and fine-grained point densification. Central to our idea is correlating implicit surface reconstruction with salient point identification, and carrying out sampling on the saliency-aware surface representation. Motivated by this, we introduce a salient point detector along with a corresponding implicit surface-based interpolator, and a geometry filter, upon which we develop a three-stage architecture consisting of a pre-trained saliency guidance block, a saliency-aware enhancer, and an upsampler. The guidance block captures meaningful shape patterns to prevent detail loss and incomplete recovery, while the enhancer facilitates detail enhancement in complex salient regions. These components are integrated with the upsampler to generate dense results that accurately retain both global shape and meticulous structures. In comparison with state-of-the-art methods, our model significantly improves the upsampling quality. Extensive experiments conducted on various datasets, comprising both synthetic and real-world captured shapes, demonstrate the flexibility and availability of our method in processing the point clouds across different scales and distributions.

Abstract:
Atmospheric blocking events are quasi-stationary high-pressure systems that disrupt the typical paths of polar and subtropical air currents, often producing prolonged extreme weather events such as summer heat waves or winter cold spells. Despite their critical role in shaping mid-latitude weather, accurately modeling and analyzing blocking events in long meteorological records remains a significant challenge. To address this challenge, we present anuncertainty visualization framework for detecting and characterizing atmospheric blocking events. First, we introduce a geometry-based detection and tracking method, evaluated on both pre-industrial climate model simulations (UKESM) and reanalysis data (ERA5), which represent historical Earth observations assimilated from satellite and station measurements onto regular numerical grids using weather models. Second, we propose a suite of uncertainty-aware summaries: contour boxplots that capture representative boundaries and their variability, frequency heatmaps that encode occurrences, and 3D temporal stacks that situate these patterns in time. Third, we demonstrate our framework in a case study of the 2003 European heatwave, mapping the spatiotemporal occurrences of blocking events using these summaries. Collectively, these uncertainty visualizations reveal where blocking events are most likely to occur and how their spatial footprints evolve over time. We envision our framework as a valuable tool for climate scientists and meteorologists: by analyzing how blocking frequency, duration, and intensity vary across regions and climate scenarios, it supports both the study of historical blocking events and the assessment of scenario-dependent climate risks associated with changes in extreme weather linked to blocking.

Abstract:
Visual cluster analysis is a central task to explore multidimensional data. Dimensionality Reduction (DR) techniques support this task by spatializing multidimensional (MD) data similarities as point patterns in scatterplots. However, unavoidable false and missing neighbor distortions limit their accuracy. For instance, false neighbors make truly separated data clusters appear to overlap in the layout, while missing neighbors split true clusters into falsely separated groups. In general, both types of distortions exist in DR layouts except for orthogonal linear projections (OLP) that only generate false neighbors. In this work, we propose Isometric Seriation-based Dimensionality Reductions (ISilDR) that provably generate at most missing neighbors. We study how ISilDR and OLP together could be leveraged to discover true MD clusters. An ISilDR first creates a seriation of the MD data points, i.e., an ordering along a one-dimensional projection axis, and then each pair of consecutive points along this axis is spaced by their MD distance. An mD ISilDR can be obtained by combining m 1D ISilDRs. We study the theoretical and empirical characteristics of different variants of ISilDRs and OLPs and propose a systematic and formal analysis based on ε-neighborhood graphs. From there, we derive rules to discover cluster patterns in MD data from interactive linking of ISilDR and OLP coordinated layouts. We then conduct case studies and illustrate scenarios for trustworthy visual cluster analysis using a combination of ISilDR and other classical DR techniques.

Abstract:
Modern simulations and observations generate vast amounts of data, fueling a growing interest in replacing discrete data with continuous surrogate models, such as functional models and implicit neural networks, to enhance data storage, transfer, analysis, and visualization. To that end, kernel regression is a well-known class of non-parametric techniques useful for surrogate modeling. In this paper, we propose a new framework to use coresets for kernel regression-a small dataset that is used as a proxy for the original data-in scientific visualization. Using kernel regression as a surrogate for scientific data, we construct an optimized coreset that is both compact and highly accurate, reducing errors from randomly-sampled and grid-based coresets by orders of magnitude. We evaluate our framework on large spatial datasets and demonstrate that it incurs negligible error while preserving the underlying topological features.

Abstract:
Recent advances in leveraging large-scale Internet photo collections for 3D reconstruction have enabled immersive virtual exploration of landmarks and historic sites worldwide. However, existing methods primarily focus on visual appearance reconstruction, often overlooking the interactive semantic understanding of these 3D scenes (e.g., identifying specific building parts or scene details), which remains largely confined to browsing static text-image pairs. Therefore, can we draw inspiration from 3D in-the-wild reconstruction techniques and use unconstrained photo collections to create an immersive approach for comprehensive 3D scene understanding beyond mere visual appearance? To this end, we extend language embedded 3D Gaussian splatting (3DGS) and propose a novel framework for open-vocabulary scene understanding from unconstrained photo collections. Specifically, we first render multiple appearance images from the same viewpoint as the unconstrained image with the reconstructed radiance field, then extract multi-appearance CLIP features and two types of language feature uncertainty maps—transient and appearance uncertainty—derived from the multi-appearance features to guide the subsequent optimization process. Next, we propose a transient uncertainty-aware autoencoder, a multi-appearance language field 3DGS representation, and a post-ensemble strategy to effectively compress, learn, and fuse language features from multiple appearances. Finally, to quantitatively evaluate our method, we introduce PT-OVS, a new benchmark dataset for assessing open-vocabulary segmentation performance on unconstrained photo collections. Experimental results show that our method outperforms existing methods, delivering accurate open-vocabulary segmentation and enabling applications such as interactive roaming with open-vocabulary queries, architectural style pattern recognition, and 3D scene editing. Visit our project page at Project Page.

Abstract:
This paper explores finger pressure as a continuous implicit input modality to enhance interaction precision in virtual reality (VR). While motion controllers are widely adopted, their limitations in delicate operations remain a critical challenge. We investigate whether finger pressure signals from conventional VR controllers could offer advantages over traditional kinematic metrics for precision interaction.Through empirical studies, we demonstrate a robust relationship between pressure dynamics and task precision requirements, leading to a lightweight sigmoid-based model that leverages detected pressure to infer desired control granularity. In a comparative evaluation of video-scrubbing tasks, our adaptive method outperforms static sensitivity baselines in both task performance and subjective preference, without elevating cognitive load. Further validation via a VR sketching application demonstrates that our technique maintains task performance while reducing mental demand compared to manual control. Our findings reveal the untapped potential of pressure-based input to bridge coarse and fine-grained VR interactions, offering a path toward more versatile and intuitive input systems.

Abstract:
Augmented reality (AR) assembly guidance systems can help users quickly master the assembly process of unfamiliar objects. However, it is very difficult for practical use without a thorough understanding of users' intentions. Furthermore, existing approaches still struggle to handle complex hand-part interactions and nonlinear assembly steps, and achieving intuitive and real-time guidance remains challenging. To address these issues, we propose an intention-driven camera-projector AR assembly guidance system (IntentionAR) that integrates online user intention inference with a finite-state assembly machine. The intention module recognizes interaction actions and infers the target object, enabling feedback before assembly begins, while the finite state machine (FSM) manages step progression. Using a camera-projector device, the system highlights candidate parts and, based on inferred intention, flags correct/incorrect selections to provide real-time guidance. As the assembly progresses, the models used for spatial registration and visualization switch dynamically to accommodate the nonlinear workflow. Experiments and user studies show that the system delivers a more robust AR interaction experience and improves assembly efficiency. A free copy of this paper and all supplemental materials, as well as project assets and source code, will be available at the project website.

Abstract:
The walking excavator is a highly complex construction machine, characterized by challenges in control, instruction, and operational safety. In this work, we implemented a virtual reality (VR) walking excavator simulator that integrates real hardware components with kinematic physics simulation. Using the simulator, we conducted two user studies. The first involved 45 participants and examined the effects of user perspectives on performance and preference. The second involved 12 participants and investigated the influence of video passthrough of the controllers. Results show that the choice of perspective should be task-dependent. Third-person view outperformed first-person view in the undercarriage task in which complex leg movements are needed. It supported better planning and allowed complete body visualization. However, it did not reduce motion sickness and was less preferred than first-person view for training. From the second user study, passthrough significantly improved controller learning compared to animated controllers, with participants making fewer errors when using the passthrough design. Based on these findings, we provide design recommendations for user perspectives and controller representation in VR-based training simulators.

Abstract:
Holographic displays offer the promise of realistic 3D visualization for virtual and augmented wearable solutions. Nevertheless, existing computer-generated holography (CGH) methods often struggle with either a high computational burden or limited display realism. While the emerging cloud-edge computing mechanism can enable the real-time streaming of holograms, classic image compression techniques struggle to efficiently encode and decode the substantial high-frequency information inherent in hologram data. In light of these challenges, we present a display-aware and lightweight CGH framework, leveraging implicit neural representations (INRs) and camera-calibrated wave propagation, to generate and compress high-fidelity phase-only holograms. Specifically, our approach interprets hologram generation as a continuous function approximation problem, enabling the network, with reduced parameters, to effectively learn the inherent periodicity and high-frequency components of 2D and 3D hologram data. To enable efficient deployment, we further incorporate quantization-aware training, followed by entropy coding. Experimental results evaluated on an unfiltered holographic display prototype demonstrate that the proposed INR-CGH retains image quality comparable to that of existing optimization-based methods in both 2D and 3D scenarios. In addition, our compact INR representation achieves up to 11× compression rate with minimal quality degradation and can be further reduced via quantization-aware training. The resulting model enables ≥250 fps in decoding speed, paving the way towards edge holography.

Abstract:
Our ability to perceive and navigate the spatial world is a cornerstone of human experience, relying on the integration of visual and auditory cues to form a coherent sense of depth and distance. In stereoscopic 3D vision, depth perception requires fixation of both eyes on a target object, which is achieved through vergence movements, with convergence for near objects and divergence for distant ones. In contrast, auditory cues provide complementary depth information through variations in loudness, interaural differences (IAD), and the frequency spectrum. We investigate the interaction between visual and auditory cues and examine how contradictory auditory information can overdrive visual depth perception in virtual reality (VR). When a new visual target appears, we introduce a spatial discrepancy between the visual and auditory cues: the visual target is shifted closer to the previously fixated object, while the corresponding sound localization is displaced in the opposite direction. By integrating these conflicting cues through multimodal processing, the resulting percept is biased toward the intended depth location. This audiovisual fusion counteracts depth compression, thus reducing the required vergence magnitude and enabling faster gaze retargeting. Such audio-driven depth enhancement may further help mitigate the vergence-accommodation conflict (VAC) in scenarios where physical depth must be compressed. In a series of psychophysical studies, we first assess the efficiency of depth overdriving for various VR-relevant combinations of initial fixations and shifted target locations, considering different scenarios of audio displacements and their loudness and frequency parameters. Next, we quantify the resulting speedup in gaze retargeting for target shifts that can be successfully overdriven by sound manipulations. Finally, we apply our method in a naturalistic VR scenario where user interface interactions with the scene show an extended perceptual depth.

Abstract:
Visual perception plays a critical role in detecting changes within immersive Virtual Reality (VR) environments. However, as visual complexity increases, perceptual performance declines, making it more difficult to detect changes quickly and accurately. This study examines how visual features, known for facilitating preattentive processing, impact a change detection task in immersive 3D environments, with a focus on visual complexity, object attributes, and spatial proximity. Our results demonstrate that preattentive processing enhances change detection, particularly when the altered object is spatially isolated and not perceptually grouped with similar surrounding objects. Changes to isolated objects were detected more reliably, suggesting that perceptual isolation reduces cognitive load and draws more attention. Conversely, when a changed object was surrounded by visually similar elements, participants were less likely to detect the change, indicating that perceptual grouping hinders individual object recognition in complex scenes. These results provide guidelines for designing VR applications that strategically utilize spatial isolation and visual features to improve the user experience.

Abstract:
Correctly estimating the surrounding illumination is essential for creating visually coherent Mixed Reality (MR) experiences. The most accurate results can be achieved by utilizing a light probe, a dedicated object with known reflectance parameters that is placed into the scene. However, the need for a dedicated object placed in the area where the illumination is estimated presents a severe limitation. Building on the increasing popularity of gestural interaction in MR, we present HandLight, an approach to estimating the illumination from the user's hands during interaction. Contrary to static light probes, HandLight does not require preparation of the environment and generates an atlas of light probes while the user moves in the world, thus reflecting variable illumination. Our system utilizes a neural network that learns the environment lighting from images of the hand. We train the network on a dataset depicting three common gestures (pinch, fist, bloom) under varying light conditions. We show that our approach can provide believable illumination estimations for a variety of illuminations on a dataset of real hand images.

Abstract:
Networked virtual reality (NVR) whiteboards are increasingly important for enabling geographically dispersed users to engage in real-time idea sharing, collaborative design, and discussion. However, latency caused by network limitations, rendering delays, or synchronization issues can significantly degrade the Quality of Experience (QoE) in whiteboard collaboration. To systematically investigate the impact of latency, this study classified QoE into pragmatic and hedonic aspects, each comprising multiple sub-dimensions. Controlled experiments were conducted to identify the sub-dimensions most affected by latency, which were then adopted as the primary QoE indicators, with the aim of uncovering the processes and mechanisms through which latency shapes QoE. Building on this, we further examined how these impacts vary across different collaboration modes, namely sequential collaboration (SC) for structured design workflows and free collaboration (FC) for open discussion. We also compared two VR whiteboard types, one with avatars (VR+) and the other without avatars (VR), and included a traditional PC-based whiteboard as a baseline. This multidimensional design enables a comprehensive evaluation of latency's impact on QoE across collaboration modes and platforms, providing practical guidance for optimizing NVR whiteboard systems under real-world network and system constraints.

Abstract:
Human trust in machines is critical for effective human-machine interaction in virtual reality (VR). Prior work defined a three-layered framework of such trust but also indicated two deficiencies. Firstly, there is an absence of a model with metrics spanning all layers to objectively capture fluctuations of the trust (trust dynamics) in real time. Secondly, there is an inadequate consideration of human and machine reliability for the trust. Herein, this study proposed a trust model by defining metrics for all the layers and evaluated this model by considering human and machine reliability. Using objective and subjective data, the evaluation was based on two VR use-cases. The outcomes of the evaluation confirmed the pertinence of the model to capture trust dynamics in the presence of human and machine reliability. The objective data was notably more sensitive to capturing trust dynamics than the subjective counterpart. The model could enable designing trustworthy and adaptive VR.

Abstract:
The task of surface reconstruction from point clouds is to produce high-quality meshes using sampled 3D points (no images available). Traditional methods primarily focus on geometric accuracy but often produce meshes without texture colors. In this paper, we present a brand-new perspective in point cloud reconstruction task—Imagining points as more informative Gaussian splats and obtaining colored surfaces through free-form Gaussian-rendering reconstruction. We train a universal Point-to-Gaussian model to infer the attributes of Gaussian splats for any given pointcloud with merely point coordinates (and color optionally) as input, without requiring any image. Significant technical designs are applied on initialization, regularization and loss functions, making the whole learning process stable. The inferred Gaussian splats can faithfully recover the original appearance of objects or scenes (capable of quick rendering from any viewpoint, like human’s imagination ability), meanwhile closely adhering to the input shape. After obtaining a sufficient number of virtually rendered images and depth maps, we employ the truncated signed distance function (TSDF) fusion to get the reconstruction results, producing high-quality and colored meshes. Extensive experiments demonstrate that our approach surpasses state-of-the-art methods in surface reconstruction metrics while maintaining high efficiency and simplicity.

Abstract:
Light field (LF) displays address the mismatch in focus cues present in traditional displays by triggering natural defocus blur and enabling motion parallax. They rely on geometrical optics, displaying rays from multiple angles of view. LF path tracing is computationally expensive for real-time applications, since it requires rendering multiple views. To reduce this computational complexity, spatially reprojecting pixels between views is commonly performed. Reusing pixels that are already rendered is cheaper than path tracing additional ones. However, when occluded areas are uncovered in some views, reprojection is not possible, creating holes in these views. Filling-in the holes requires extra path tracing computation. This paper investigates scalable hole-filling strategies for LF path tracing, using multiple GPUs to reach real-time performance. We propose an algorithm search optimization procedure to determine whether a specific assignment algorithm can be generalized across scenes, using the hole-filling time as a minimization function. In addition, we introduce DaSH (Discarded and Subsampled path traced Hole-filling), a novel method that reduces computation and divergence overhead in fixed-size hardware thread-blocks. Based on local pixel sparsity within pixel patches, DaSH adaptively subsamples and discards hole-filling rays. Our evaluation demonstrates that DaSH achieves significant performance gains while preserving the visual and structural quality of refocused light field images at the retina plane. The experiments demonstrate an average speedup factor of 1.8×1.8× for DaSH, compared to prior work, in a multi-GPU rendering system.

Abstract:
Datasets with large heterogeneity in data density are prone to result in visualizations with unutilized whitespace and overplotting. Sequential data, such as in genomics or electronic health records, often has such density distributions, containing long empty stretches between data-rich regions. For these data types, the exact distance is not necessarily the central feature, and plotting along a linear axis causes occlusion of the actual datapoints. In this work, we evaluate content-driven axis compression for minimizing whitespace and reducing occlusion while maintaining distance information. The axis is compressed based on the density of the data, and it requires minimal a priori knowledge of the data distribution. We present a framework for reasoning about the benefits and drawbacks of using axis compression. The framework includes reasoning about the separability of items and clusters, suggests careful consideration of how the position visual channel is utilized, and highlights the influence of the usage context for which the visualization is designed. We conducted a study to evaluate the strengths and weaknesses of axis compression and found that it leads to improved item visibility while maintaining high accuracy, also for tasks targeted at understanding the data distribution.

Abstract:
Recently, 3D Gaussian Splatting (3DGS) has emerged as a powerful technique for reconstructing high-quality, photorealistic 3D representations of real-world scenes. However, editing 3DGS remains more challenging than mesh-based approaches as it lacks explicit geometry and requires view-consistent updates across Gaussians. Previous approaches have relied primarily on text-driven generative models for 3D Gaussian editing, limiting direct control over specific visual appearances. In this study, we propose a texture editing framework for 3DGS that leverages only a single user-modified image. Our approach maintains coherence across multiple views while accounting for the corresponding lighting adjustments in the edited region. We introduce three key techniques to achieve this: (1) aligned edit propagation, which transfers local edits from the reference view to other viewpoints; (2) mask-based filtering, which restricts modifications to relevant Gaussians and prevents unintended changes; and (3) opacity-based selection, which identifies Gaussians with the most significant visual impact on the edited texture. Through both qualitative and quantitative evaluations on synthetic and real-world datasets, we demonstrate that our method achieves more precise and spatially controllable 3DGS editing than existing techniques. We expect these techniques to pave the way for more intuitive 3D Gaussian Splatting editing pipelines and inspire future research.

Abstract:
Creating realistic and fully detailed 3D human models using a minimal number of views has long been a challenging goal in 3D human reconstruction. Reconstructing a realistic human model from only two images (front and back) is particularly difficult due to the limited 3D information available, leading to two major challenges: (1) it is difficult to establish spatial consistency for reconstruction due to the lack of sufficient images for reliable matching, and (2) incomplete field of view results in missing color information. To address these challenges, we propose DuetGS, a novel pipeline that divides the reconstruction process into two stages: geometry reconstruction and color reconstruction. For geometry reconstruction, we employ a data-driven neural network to recover a full-body mesh from the front and back images, providing the spatial positioning for Gaussians. For color reconstruction, we adapt Gaussian Splatting and integrate our proposed unsupervised color propagation method to establish the color details of the Gaussians. Furthermore, our Gaussians are directly mapped to the mesh, allowing us to control their rotation and translation through mesh manipulation. This mapping ensures compatibility with various animation techniques. Extensive experiments on the THUman, CustomHumans, and PeopleSnapshot datasets demonstrate that our approach outperforms existing methods in terms of reconstruction accuracy and visual quality.

Abstract:
Due to substantial differences in appearance and movement between cartoon faces and real human faces, directly applying existing speech-driven speaker generation techniques to cartoon face images often fails to accurately capture key points, resulting in distorted expressions and animation artifacts. To address these challenges, this article proposes a speech-driven cartoon face animation generation model called CartoonTalk, which integrates 2D cartoon key points with 3D motion coefficients. The model comprises three modules: face annotation, motion extraction, and 3D facial rendering. The face annotation module first annotates a cartoon image, capturing its 2D key points. Subsequently, the motion extraction module derives 3D motion coefficients for head and facial expressions from speech data. Finally, the 3D facial rendering module uses these 2D and 3D features to generate the animated cartoon face. Quantitative and qualitative evaluations demonstrate that our model significantly outperforms existing approaches in both lip-sync accuracy and visual quality, establishing new state-of-the-art benchmarks on cartoon datasets.

Abstract:
Efficient and realistic relighting of complex scenes with unknown illumination remains a crucial but challenging task. Recent advancements in 3D Gaussian Splatting (3DGS) have shown impressive object-level relighting. However, they still struggle with complex real-world scenes, mainly due to the challenges of accurately decoupling intricate geometry, materials, and lighting using concise 3D Gaussian primitives. In this paper, we propose a new Geometrically-Aligned Neural Gaussian Splatting (GANG) method, which performs efficient physically based rendering (PBR) directly on anchor-based relightable neural Gaussians. Our key idea is to regularize the decoded neural Gaussians geometrically aligned with the latent signed distance field (SDF) surface spawned from anchors using a differentiable implicit indicator function (IIF) solver. It brings effective geometric association to accurate decoupling of materials and lighting for efficient and realistic relighting of complex scenes. Furthermore, we propose a locally consistent geometry regularization to guide more concise neural Gaussian learning with a hybrid lighting model, which combines position-learnable spherical Gaussians (SGs) and an environment map, allowing accurate modeling of both local and global illumination. Experimental results on public datasets demonstrate that GANG consistently outperforms previous PBR methods in material decomposition and relighting quality, while representing complex scenes with concise anchors. To the best of our knowledge, GANG is a new state-of-the-art 3DGS method for realistic relighting, enabling efficient rendering and flexible editing materials and illumination, especially for complex scenes.

Abstract:
The recent development of 3D Gaussian splatting (3DGS) has led to great interest in 4D dynamic spatial reconstruction. Existing approaches mainly rely on full-length multi-view videos, while there has been limited exploration of online reconstruction methods that enable on-the-fly training and per-timestep streaming. Current 3DGS-based streaming methods treat the Gaussian primitives uniformly and constantly renew the densified Gaussians. Thus, they overlook the difference between dynamic and static features and neglect the temporal continuity of the scene. To address these limitations, we propose a novel pipeline for iterative streamable 4D dynamic spatial reconstruction. It comprises three stages: a selective inheritance stage that retains priors from previous timesteps to preserve the temporal continuity, a dynamics-aware shift stage that distinguishes dynamic and static primitives and employs distinct strategies to optimize their movements, and an error-guided densification stage that efficiently identifies Gaussians requiring densification to accommodate emerging objects. Our method achieves state-of-the-art performance in online 4D reconstruction, demonstrating compact storage, the fastest on-the-fly training speed, and superior representation quality.

Abstract:
Controllable image-to-video (I2V) generation transforms a reference image into a coherent video guided by user-specified control signals. While precise control over camera motion, object motion, and lighting is essential for high-fidelity creation, existing methods often treat these factors independently. This overlooks the physical coupling among viewpoint, geometry, and illumination in dynamic scenes, leading to visual inconsistencies such as mismatched shadows and perspective drift under simultaneous changes. We present VidCRAFT3, a unified and flexible I2V framework that explicitly models cross-factor interactions among geometry, motion, and illumination, enabling both independent and joint control over camera motion, object motion, and lighting direction. Image2Cloud provides explicit 3D geometric priors for accurate camera motion control. ObjMotionNet encodes sparse object trajectories into multi-scale motion features to guide realistic object motion. A Spatial Triple-Attention Transformer integrates lighting direction through lighting cross-attention for consistent relighting. To address the scarcity of jointly annotated data, we construct the VideoLightingDirection (VLD) dataset with accurate per-frame lighting direction annotations, and introduce a three-stage progressive training strategy that enables robust learning without fully joint annotations. Extensive experiments demonstrate that VidCRAFT3 achieves state-of-the-art performance in control precision and visual coherence across diverse scenarios.

Abstract:
We have recently seen some progress in the current field of human-human interaction generation. However, directly generating complex two-person interactive motions remains a significant challenge. Meanwhile, these models typically employ two independent timelines when generating motions for interactive scenarios involving two individuals. This design overlooks the temporal dependencies between motions at each timestep and fails to account for the roles of active and reactive participants during the generation process, often resulting in unrealistic and unnatural motions. In this work, we propose HiTMM, a novel framework for Human interaction generation based on Temporal Masked Modeling. HiTMM first decomposes the human interaction into two separate single-person motions. Individual motions within the interaction belong to the same type, enabling them to be mapped to a shared latent space through a coarse-to-fine approach that produces multi-layer discrete tokens. We then arrange all tokens of the two interacting individuals along a shared timeline. Subsequently, we employ a masked transformer and a residual transformer to model the base-layer and rest-layer motion tokens. Both the base-layer and rest-layer motion tokens are arranged along a single timeline, allowing the model to explicitly capture the temporal order and initiating role embedded in the sequence, where the first individual’s motion initiates the interaction. Note that, our model utilizes a shared temporal representation, making it capable of performing temporal editing on specific regions within human interaction sequences. Experimental results show that our model achieves an FID of 5.017 on the InterHuman dataset, surpassing the current state-of-the-art model (vs 5.154 for InterMask), and an FID of 0.373 on the InterX dataset (vs 0.399 for InterMask).

Abstract:
Visual analysis of spatiotemporal density distributions is crucial for understanding spatiotemporal dynamics. However, existing methods suffer from visual occlusion and information loss when simultaneously displaying multiple density distributions. We present Strunkmap as an abstract approach to address these challenges. We introduce anisotropic kernel density estimation to enhance the accuracy of density generation. We extract the trunks of density distributions to identify the overall spatial patterns. Path scanning and trunk-outline matching strategies are employed to preserve local spatial structure. We design a stacked trunk plot that enables lossless density representation while conserving substantial screen space. Based on the visual design, Strunkmap integrates multiple heatmaps within a single map to effectively display temporal evolution of density distributions without visual occlusion. Ablation studies and comparative experiments validate the superiority of Strunkmap in accuracy and efficiency for hotspot identification and trend exploration. Theoretical analysis demonstrates Strunkmap’s scalability, which we further verify through large-scale spatiotemporal data visualization. Color encoding schemes and scaling ratios are discussed to illustrate the flexibility. Our evaluations with user feedback demonstrate that Strunkmap is a viable solution with significant potential to real-world applications.

Abstract:
Recent advances in generative models have yielded impressive progress on motion in-betweening, allowing for more complex, varied, and realistic motion transitions. However, recent methods still exhibit noticeable limitations in preserving keyframe information and ensuring motion continuity. In this paper, we propose a novel pipeline and sampling optimization strategy for latent diffusion models (LDM) based on motion implicit neural representations (INR). By establishing a mapping between INR and sparse spatial or temporal information within latent diffusion, our model can sample the INR parameters from extremely sparse and ambiguous keyframe data and reconstruct plausible and smooth motions from the manifold. Our experiments demonstrate the superior performance of our model, which significantly improves motion generation quality in scenarios with few keyframes while ensuring both keyframe accuracy and diversity of in-between motions.

Abstract:
We present a new nonlinear dimensionality reduction method, MAPLE, that enhances UMAP by improving manifold modeling. MAPLE employs a self-supervised learning approach to more efficiently encode low-dimensional manifold geometry. Central to this approach are maximum manifold capacity representations (MMCRs), which help untangle complex manifolds by compressing variances among locally similar data points while amplifying variance among dissimilar data points. This design is particularly effective for high-dimensional data with substantial intra-cluster variance and curved manifold structures, such as biological or image data. Our qualitative and quantitative evaluations demonstrate that MAPLE can produce clearer visual cluster separations and finer subcluster resolution than UMAP while maintaining a tractable computational cost.

Abstract:
Generative AI techniques are revolutionizing the creation of 3D and immersive content, yet challenges remain, such as achieving precise user control in 3D generation. Current text-to-3D and image-to-3D pipelines often produce outputs that deviate from user expectations, lacking the ability to refine or correct generated models effectively. To address these limitations, we propose Imagin3D, a novel human-in-the-loop (HITL) system that integrates Multimodal Large Language Models to enhance the controllability and adaptability of 3D content generation. Imagin3D leverages a Multi-View Question Answering module to evaluate the consistency of generated views with user-provided textual descriptions, enabling iterative refinement through guided inpainting while preserving multi-view consistency. This allows users to co-create 3D models, which are then synthesized into a final 3D asset using Neural Rendering. We validate Imagin3D through extensive quantitative evaluations and a comprehensive user study, demonstrating its effectiveness in improving usability, accuracy, and user satisfaction in interactive 3D generation tasks. Our results highlight the potential of HITL approaches to bridge the gap between AI-generated outputs and user intent, paving the way for more accessible and user-centered 3D generation workflows.

Abstract:
Articles containing geographic information are widely distributed and commonly used in daily life, frequently incorporating geographic visualizations as illustrations. However, the creation of such articles remains cumbersome, necessitating authors to switch between authoring text and illustrations, thereby disrupting immersive writing. Our interviews corroborated this observation and revealed the primary challenge in the traditional process stems from the low synchronization frequency between text and geographic visualizations during creation, coupled with weak visual links, forcing users to mentally maintain this synchronization and thereby increasing their cognitive burden. In response, we developed GeoAuthor, which facilitates the interactive creation of geographic articles by automatically synchronizing text creation with geographic visualizations with rich visual links. This bidirectional approach ensures that the written content and visual representations remain consistent and mutually informative throughout the creation process. Our evaluation demonstrated the efficacy of GeoAuthor, indicating its capacity to streamline the process of creating geographic articles.

Abstract:
This literature review examines the use of virtual reality in ergonomic risk assessment in workplace design using a rigorous methodology to evaluate literature from seven databases and journals. The review identified 4,377 records, of which nineteen used virtual reality to simulate complex work environments, such as manufacturing floors and office spaces, utilizing motion tracking and risk assessment methods, including the Ergonomic Assessment Worksheet (EAWS) or CUELA (German acronym for Computer-aided recording and long-term analysis of musculoskeletal system loads). The research suggests that integrating virtual reality with established ergonomic assessment methods holds promise for enhancing their efficacy. However, it also highlights the limitations of current virtual reality simulation tools in implementing realistic interactions with the working environment. The importance of this review lies in its capacity to synthesize and structure a previously fragmented field, providing researchers and practitioners with a comprehensive overview of current approaches, challenges, and opportunities in VR-based ergonomic risk assessment. It concludes that virtual reality has high potential but requires further research to fully realize its benefits in designing ergonomic workplaces, conducting thorough ergonomic evaluations, and being integrated into occupational health and safety on a large scale.

Abstract:
Talking head generation is increasingly important in virtual reality (VR), especially for social scenarios involving multi-turn conversation. Existing approaches face notable limitations: mesh-based 3D methods can model dual-person dialogue but lack realistic textures, large-model-based 2D methods produce natural appearances but incur prohibitive computational costs. Recently, 3D Gaussian Splatting (3DGS)-based methods achieve efficient and realistic rendering but remain speaker-only and ignore social relationships. We introduce RSATalker, the first framework that leverages 3DGS for realistic and socially-aware talking head generation, with support for multi-turn conversation. Our method first drives mesh-based 3D facial motion from speech, then binds 3D Gaussians to mesh facets to render high-fidelity 2D avatar videos. To capture interpersonal dynamics, we propose a socially-aware module that encodes social relationships, including blood and non-blood as well as equal and unequal, into high-level embeddings through a learnable query mechanism. We design a three-stage training paradigm and construct the RSATalker dataset with speech-mesh-image triplets annotated with social relationships. Our method supports applications such as VR telepresence, social VR, and embodied conversational agents. The socially-aware conditioning can also be extended to other human motion generation tasks. Extensive experiments demonstrate that RSATalker achieves state-of-the-art performance in both realism and social awareness. The code and dataset will be released.

Abstract:
With the increasing popularity of virtual reality (VR) applications, allowing interaction with non-immersed spectators has become an important supporting functionality. To this end, a variety of VR streaming systems focus on allowing remote spectating experiences. However, such systems lack mechanisms that allow spectators to assist or explore inside VR applications alongside the VR players. To address, we explore AccompliceVR (AVR), a novel streaming and remote collaboration system. Using built-in OpenVR functionality, AVR introduces avatar representations of spectators overlaid into the normal stereo views of unmodified VR applications. AVR also enhances the remote spectator experience by stitching frames of video from the VR player to reconstruct a pseudo-panoramic view. This allows them to move around and point at the environment with an avatar. We first evaluated that AVR is compatible with a selection of the most popular VR games. The system is then validated through a paired user study with 30 participants using a mixed-design pattern. Participants played two well-known VR games (as ‘pilot’) and were assisted by their partner in the pair (as ‘copilot’). The results support that AVR shows significant improvement in communication between pilots and copilots and enhances the pilot's understanding of instructions from their copilot.

Abstract:
In traditional visual analysis, brushing and linking is commonly used to visually connect multiple views using highlighting techniques. However, brushing and linking has rarely been used in situated analytics, which uses visualizations to analyze data in the context of physical referents. In situated analytics, data representations must be visually linked to real-world objects. Previous work has assessed situated brushing and linking in a virtual reality simulation of a supermarket scenario. Here, we replicate and extend the previous approach by studying brushing and linking in an actual physical space with augmented reality, while further improving the highlighting techniques. Using a video see-through display, we compare augmented reality with virtual reality. Results suggest that AR performs better in time and accuracy, but the effectiveness of the techniques varies by condition. These results provide a new framing of how the real-world stimuli matter in situated analytics.

Abstract:
Monocular dynamic video reconstruction is a typical ill-posed problem due to the limited observations and complex 3D motions. Despite the recent advances in dynamic 3D Gaussian splatting techniques, most of them still struggle with the monocular setting, since they heavily rely on geometric cues from multiple cameras or ignore the structural coherence among the optimized 3D Gaussains. To address this, we propose Hie4DGS, a novel hierarchical structure representation to model the complex dynamic motions from monocular dynamic videos. Specifically, we decompose the motions of a dynamic scene into groups of multiple structure granularities and progressively compose them to derive the motion of each 3D Gaussian. Building on this representation, we leverage hierarchical semantic segmentation to group Gaussians and initialize their motion using depth and tracking priors within each group. Additionally, we introduce a structure rendering loss that enforces consistency between the learned motion structure and semantic priors, further reducing motion ambiguity. Compared to the state-of-the-art dynamic Gaussian methods, we achieve significant improvement in rendering quality on monocular video datasets featuring complex real-world motions.

Abstract:
While diffusion-based methods have shown impressive capabilities in capturing diverse and complex hairstyles, their ability to generate consistent and high-quality multi-view outputs—crucial for real-world applications such as digital humans and virtual avatars—remains underexplored. In this paper, we propose Stable-Hair v2, a novel diffusion-based multi-view hair transfer framework. To the best of our knowledge, this is the first work to leverage multiple-view diffusion models for robust, high-fidelity, and view-consistent hair transfer across multiple perspectives. We introduce a comprehensive multi-view training data generation pipeline to generate high-quality triplet data, including bald images, reference hairstyles, and view-aligned source-bald pairs. Our multi-view hair transfer model integrates polar-azimuth embeddings for pose conditioning and temporal attention layers to ensure smooth transitions between views. To optimize this model, we design a novel multi-stage training strategy consisting of Pose-Controllable Latent IdentityNet training, Hair Extractor training, and Temporal Attention training. Extensive experiments demonstrate that our method accurately transfers detailed and realistic hairstyles to source subjects while achieving seamless and consistent results across views, significantly outperforming existing methods and establishing a new benchmark in multi-view hair transfer.

Abstract:
3D scene generation is in high demand across various domains, including virtual reality, gaming, and the film industry. Owing to the powerful generative capabilities of text-to-image diffusion models that provide reliable priors, creating 3D scenes using only text prompts has become viable, thereby significantly advancing research in text-driven 3D scene generation. Prevailing methods typically employ the diffusion model to generate an initial local image, followed by iteratively outpainting the local image to gradually generate scenes. Nevertheless, these outpainting-based approaches are prone to producing globally inconsistent results with low completeness, restricting their broader applications. To tackle these problems, we introduce HoloDreamer, a framework that begins by generating a high-definition panorama to holistically initialize the full scene, and leverages 3D Gaussian Splatting (3D-GS) for rapid 3D scene reconstruction, thereby facilitating the creation of view-consistent and fully enclosed 3D scenes. Specifically, we propose Stylized Equirectangular Panorama Generation, a pipeline that combines multiple diffusion models to enable stylized and detailed equirectangular panorama generation from complex text prompts. Subsequently, Enhanced Two-Stage Panorama Reconstruction is introduced, conducting a two-stage optimization of 3D-GS to inpaint the missing region and enhance the integrity of the scene. Comprehensive experiments demonstrated that our method outperforms prior works in terms of overall visual consistency and harmony, as well as reconstruction quality and rendering robustness when generating fully enclosed scenes.

Abstract:
3D Gaussian Splatting (3DGS) has recently attracted wide attentions in various areas such as 3D navigation, Virtual Reality (VR) and 3D simulation, due to its photorealistic and efficient rendering performance. High-quality reconstrution of 3DGS relies on sufficient splats and a reasonable distribution of these splats to fit real geometric surface and texture details, which turns out to be a challenging problem. We present GeoTexDensifier, a novel geometry-texture-aware densification strategy to reconstruct high-quality Gaussian splats which better comply with the geometric structure and texture richness of the scene. Specifically, our GeoTexDensifier framework carries out an auxiliary texture-aware densification method to produce a denser distribution of splats in fully textured areas, while keeping sparsity in low-texture regions to maintain the quality of Gaussian point cloud. Meanwhile, a geometry-aware splitting strategy takes depth and normal priors to guide the splitting sampling and filter out the noisy splats whose initial positions are far from the actual geometric surfaces they aim to fit, under a Validation of Depth Ratio Change checking. With the help of relative monocular depth prior, such geometry-aware validation can effectively reduce the influence of scattered Gaussians to the final rendering quality, especially in regions with weak textures or without sufficient training views. The texture-aware densification and geometry-aware splitting strategies are fully combined to obtain a set of high-quality Gaussian splats. We experiment our GeoTexDensifier framework on various datasets and compare our Novel View Synthesis results to other state-of-the-art 3DGS approaches, with detailed quantitative and qualitative evaluations to demonstrate the effectiveness of our method in producing more photorealistic 3DGS models.

Abstract:
Neural Radiance Fields (NeRF) have gained significant attention due to their precise reconstruction and rapid inference capabilities, making them highly promising for applications in virtual reality and gaming. However, extending NeRF’s capabilities to dynamic scenes remains underexplored, particularly in ensuring consistent and coherent reconstructions across space, time, and viewing angles. To address this challenge, we propose Scale-NeRF, a novel approach that organizes the training of dynamic NeRFs as a progressive, scale-based refinement process, grounded in hierarchical Bayesian theory. Scale-NeRF begins by reconstructing the radiance fields using coarse, large-scale frames and iteratively refines them with progressively smaller-scale frames. This hierarchical strategy, combined with a corresponding sampling approach and a newly introduced structural loss, ensures consistency and integrity throughout the reconstruction process. Experiments on public datasets validate the superiority of Scale-NeRF over traditional methods, especially in terms of the proposed metrics evaluating spatial, angular, and temporal consistency. Furthermore, Scale-NeRF demonstrates excellent dynamic reconstruction capabilities with real-time rendering, offering a significant advancement for applications demanding both high fidelity and real-time performance.

Abstract:
3D Gaussian Splatting has emerged as a promising technique for high-quality 3D rendering, leading to increasing interest in integrating 3DGS into realism SLAM systems. However, existing methods face challenges such as Gaussian primitives redundancy, forgetting problem during continuous optimization, and difficulty in initializing primitives in monocular case due to lack of depth information. In order to achieve efficient and photorealistic mapping, we propose RP-SLAM, a 3D Gaussian splatting-based vision SLAM method for monocular and RGB-D cameras. RP-SLAM decouples camera poses estimation from Gaussian primitives optimization and consists of three key components. Firstly, we propose an efficient incremental mapping approach to achieve a compact and accurate representation of the scene through adaptive sampling and Gaussian primitives filtering. Secondly, a dynamic window optimization method is proposed to mitigate the forgetting problem and improve map consistency. Finally, for the monocular case, a monocular keyframe initialization method based on sparse point cloud is proposed to improve the initialization accuracy of Gaussian primitives, which provides a geometric basis for subsequent optimization. The results of numerous experiments demonstrate that RP-SLAM achieves state-of-the-art map rendering accuracy while ensuring real-time performance and model compactness.

Abstract:
Traditional force-based graph layout models are rooted in virtual physics, while criteria-driven techniques position nodes by directly optimizing graph readability criteria. In this article, we systematically explore the integration of these two approaches, introducing criteria-driven force-based graph layout techniques. We propose a general framework that, based on user-specified readability criteria, such as minimizing edge crossings, automatically constructs a force-based model tailored to generate layouts for a given graph. Models derived from highly similar graphs can be reused to create initial layouts, users can further refine layouts by imposing different criteria on subgraphs. We perform quantitative comparisons between our layout methods and existing techniques across various graphs and present a case study on graph exploration. Our results indicate that our framework generates superior layouts compared to existing techniques and exhibits better generalization capabilities than deep learning-based methods.

Abstract:
Estimating the normal of a point requires constructing a local patch to provide center-surrounding context, but determining the appropriate neighborhood size is difficult when dealing with different data or geometries. Existing methods commonly employ various parameter-heavy strategies to extract a full feature description from the input patch. However, they still have difficulties in accurately and efficiently predicting normals for various point clouds. In this work, we present a new idea of feature extraction for robust normal estimation of point clouds. We use the fusion of multi-scale features from different neighborhood sizes to address the issue of selecting reasonable patch sizes for various data or geometries. We seek to model a patch feature fitting (PFF) based on multi-scale features to approximate the optimal geometric description for normal estimation and implement the approximation process via multi-scale feature aggregation and cross-scale feature compensation. The feature aggregation module progressively aggregates the patch features of different scales to the center of the patch and shrinks the patch size by removing points far from the center. It not only enables the network to precisely capture the structure characteristic in a wide range, but also describes highly detailed geometries. The feature compensation module ensures the reusability of features from earlier layers of large scales and reveals associated information in different patch sizes. Our approximation strategy based on aggregating the features of multiple scales enables the model to achieve scale adaptation of varying local patches and deliver the optimal feature description. Extensive experiments demonstrate that our method achieves state-of-the-art performance on both synthetic and real-world datasets with fewer network parameters and running time.

Abstract:
Human-human interaction generation has garnered significant attention in motion synthesis due to its vital role in understanding humans as social beings. However, existing methods typically rely on transformer-based architectures, which often face challenges related to scalability and efficiency. To address these challenges, we propose InterMamba, a novel and efficient human–human interaction generation method built on the Mamba framework, designed to capture long-sequence dependencies effectively while enabling real-time feedback. Specifically, we introduce an adaptive spatio-temporal Mamba framework that utilizes two parallel SSM branches with an adaptive mechanism to integrate the spatial and temporal features of motion sequences. To further enhance the model’s ability to capture dependencies within individual motion sequences and the interactions between different individual sequences, we develop two key modules: the self adaptive spatio-temporal Mamba module and the cross adaptive spatio-temporal Mamba module, enabling efficient feature learning. Extensive experiments demonstrate that our method achieves the state-of-the-art results on both two interaction datasets with remarkable quality and efficiency. Compared to the baseline method InterGen, our approach not only improves accuracy but also reduces the parameter size to just 66 M (36% of InterGen’s), while achieving an average inference speed of 0.57 seconds, which is 46% of InterGen’s execution time.

Abstract:
Found object drawing is a creative art form incorporating everyday objects into imaginative images, offering a refreshing and unique way to express ideas. However, for many people, creating this type of work can be challenging due to difficulties in generating creative ideas and finding suitable reference images to help translate their ideas onto paper. Based on the findings of a formative study, we propose GenFODrawing, a creativity support tool to help users create diverse found object drawings. Our system provides AI-driven textual and visual inspirations, and enhances controllability through sketch-based and box-conditioned image generation, enabling users to create personalized outputs. We conducted a user study with twelve participants to compare GenFODrawing, to a baseline condition where the participants completed the creative tasks using their own desired approaches without access to our system. The study demonstrated that GenFODrawing, enabled easier exploration of diverse ideas, greater agency and control through the creative process, and higher creativity support compared to the baseline. A further open-ended study demonstrated the system’s usability and expressiveness, and all participants found the creative process engaging.

Abstract:
Gaussian SLAM systems excel in real-time rendering and fine-grained reconstruction compared to NeRF-based systems. However, their reliance on extensive keyframes is impractical for deployment in real-world robotic systems, which typically operate under sparse-view conditions that can result in substantial holes in the map. To address these challenges, we introduce DenseSplat, the first SLAM system that effectively combines the advantages of NeRF and 3DGS. DenseSplat utilizes sparse keyframes and NeRF priors for initializing primitives that densely populate maps and seamlessly fill gaps. It also implements geometry-aware primitive sampling and pruning strategies to manage granularity and enhance rendering efficiency. Moreover, DenseSplat integrates loop closure and bundle adjustment, significantly enhancing frame-to-frame tracking accuracy. Extensive experiments on multiple large-scale datasets demonstrate that DenseSplat achieves superior performance in tracking and mapping compared to current state-of-the-art methods.

Abstract:
No-reference bitstream-layer point cloud quality assessment (PCQA) can be deployed without full decoding at any network node to achieve real-time quality monitoring. In this work, we develop the first PCQA model dedicated to Trisoup-Lifting encoded 3D point clouds by analyzing bitstreams without full decoding. Specifically, we investigate the relationship among texture bitrate per point (TBPP), texture complexity (TC) and texture quantization parameter (TQP) while geometry encoding is lossless. Subsequently, we estimate TC by utilizing TQP and TBPP. Then, we establish a texture distortion evaluation model based on TC, TBPP and TQP. Ultimately, by integrating this texture distortion model with a geometry attenuation factor, a function of trisoupNodeSizeLog2 (tNSL), we acquire a comprehensive NR bitstream-layer PCQA model named streamPCQ-TL. In addition, this work establishes a database named WPC6.0, the first PCQA database dedicated to Trisoup-Lifting encoding mode, encompassing 400 distorted point clouds with 4 geometry multiplied by 5 texture distortion levels. Experiment results on M-PCCD, ICIP2020 and the proposed WPC6.0 database suggest that the proposed streamPCQ-TL model exhibits robust and notable performance in contrast to existing advanced PCQA metrics, particularly in terms of computational cost.

Abstract:
We present HuGDiffusion, a generalizable 3D Gaussian splatting (3DGS) learning pipeline to achieve novel view synthesis (NVS) of human characters from single-view input images. Existing approaches typically require monocular videos or calibrated multi-view images as inputs, whose applicability could be weakened in real-world scenarios with arbitrary and/or unknown camera poses. In this paper, we aim to generate the set of 3DGS attributes via a diffusion-based framework conditioned on human priors extracted from a single image. Specifically, we begin with carefully integrated human-centric feature extraction procedures to deduce informative conditioning signals. Based on our empirical observations that jointly learning the whole 3DGS attributes is challenging to optimize, we design a multi-stage generation strategy to obtain different types of 3DGS attributes. To facilitate the training process, we investigate constructing proxy ground-truth 3D Gaussian attributes as high-quality attribute-level supervision signals. Through extensive experiments, our HuGDiffusion shows significant performance improvements over the state-of-the-art methods.

Abstract:
The challenge of transforming partial point clouds into complete meshes still persists, with current methods facing issues like data accessibility constraint, shape preservation failure and poor robustness on real-scan data. Drawing inspiration from the structural information of objects to enhance the completion, we introduce an innovative weakly-supervised shape completion method leveraging structural decomposition without the necessity of SDFs during training. By representing objects as abstract structural frameworks and part details, our method initiates by forecasting the structure of the input partial point clouds, and individually restore each component through part decomposition completion and generation. Extracted part details are represented in images, which are porous and incomplete. Hence, we utilize a completion network to complete such details. For multiple results generation, a diffusion-based generation network is employed to generate a variety of details for the missing areas. The predicted structure and details are subsequently converted back into meshes, yielding the complete results. Since the details are depicted in images, our approach eliminates the need for SDFs during the training phase, achieving weakly-supervision. We conduct extensive comparisons on both artificial and real-scan datasets, demonstrating an average improvement of over 38.1% compared to the prior method, and achieving SOTA performance.

Abstract:
Light field (LF) imaging is inherently constrained by the trade-off between spatial resolution and angular sampling density. To overcome this obstacle, spatial-angular super-resolution (SR) methods have been developed to achieve concurrent enhancement in both dimensions. Traditional spatial-angular SR methods treat spatial and angular SR as separate tasks, resulting in parameter redundancy and error accumulation. While recent end-to-end approaches attempt joint processing, their uniform treatment of these distinct problems overlooks critical domain-specific requirements. To address these challenges, we propose a domain-specialized framework that deploys stage-tailored strategies to satisfy domain-specific demands. Specifically, in the angular SR stage, we introduce a cross-view consistency modulation module that enhances inter-view coherence through long-range dependency modeling of angular features. In the spatial SR stage, we propose a detail-aware state space model to reconstruct fine-grained detail. Finally, we develop a cross-domain integration module that explores spatial-angular correlations by fusing multi-representational features from both domains to foster synergistic optimization. Experimental results on public LF datasets demonstrate substantial improvements over state-of-the-art methods in both qualitative and quantitative comparisons, with approximately 50% fewer model parameters compared to competing methods.

Abstract:
Brushing is a common interaction technique in 2D scatterplots, allowing users to select clustered points within a continuous, enclosed region for further analysis or filtering. However, applying conventional brushing to 2D representations of multidimensional (MD) data, i.e., Multidimensional Projections (MDPs), can lead to unreliable cluster analysis due to MDP-induced distortions that inaccurately represent the cluster structure of the original MD data. To alleviate this problem, we introduce a novel brushing technique for MDPs called Distortion-aware brushing. As users perform brushing, Distortion-aware brushing correct distortions around the currently brushed points by dynamically relocating points in the projection, pulling data points close to the brushed points in MD space while pushing distant ones apart. This dynamic adjustment helps users brush MD clusters more accurately, leading to more reliable cluster analysis. Our user studies with 24 participants show that Distortion-aware brushing significantly outperforms previous brushing techniques for MDPs in accurately separating clusters in the MD space and remains robust against distortions. We further demonstrate the effectiveness of our technique through two use cases: (1) conducting cluster analysis of geospatial data and (2) interactively labeling MD clusters.

Abstract:
Automatic furniture layout is long desired for convenient interior design. Leveraging the remarkable visual reasoning capabilities of multimodal large language models (MLLMs), recent methods address layout generation in a static manner, lacking the feedback-driven refinement essential for interactive user engagement. We introduce Chat2Layout, a novel interactive furniture layout generation system that extends the functionality of MLLMs into the realm of interactive layout design. To achieve this, we establish a unified vision-question paradigm for in-context learning, enabling seamless communication with MLLMs to steer their behavior without altering model weights. Within this framework, we present a novel training-free visual prompting mechanism. This involves a visual-text prompting technique that assist MLLMs in reasoning about plausible layout plans, followed by an Offline-to-Online search (O2O-Search) method, which identifies the minimal set of informative references to provide exemplars for visual-text prompting. By employing an agent system with MLLMs as the core controller, we enable bidirectional interaction. The agent not only comprehends the 3D environment and user requirements through linguistic and visual perception but also plans tasks and reasons about actions to generate and arrange furniture within the virtual space. Furthermore, the agent iteratively updates based on visual feedback from execution results. Experimental results demonstrate that our approach facilitates language-interactive generation and arrangement for diverse and complex 3D furniture.

Abstract:
Text-driven style transfer methods leveraging diffusion models have shown impressive creativity, yet they still face challenges in maintaining consistent structure and content preservation. Existing methods often directly concatenate the content and style prompts for a prompt-level style injection. However, this coarse-grained style injection strategy inevitably leads to structural deviations in the stylized images. This poses a significant obstacle for professional artists and creators seeking precise artistic editing. In this work, we strive to attain a harmonious balance between content preservation and style transformation. We propose Adaptive Style Incorporation (ASI), to achieve fine-grained feature-level style incorporation. It consists of the Siamese Cross-Attention (SiCA) to decouple the single-track cross-attention to a dual-track structure to obtain separate content and style features, and the Adaptive Content-Style Blending (AdaBlending) module to couple the content and style information from a structure-consistent manner. Experimentally, our method exhibits much better performance in both structure preservation and stylized effects.

Abstract:
Maps are essential to news media as they provide a familiar way to convey spatial context and present engaging narratives. However, the design of journalistic maps may be challenging, as editorial teams need to balance multiple aspects, such as aesthetics, the audience's expected data literacy, tight publication deadlines, and the team's technical skills. Data journalists often come from multiple areas and lack a cartography, data visualization, and data science background, limiting their competence in creating maps. While previous studies have examined spatial visualizations in data stories, this research seeks to gain a deeper understanding of the map design process employed by news outlets. To achieve this, we strive to answer two specific research questions: what is the design space of journalistic maps? and how do editorial teams produce journalistic map articles? To answer the first one, we collected and analyzed a large corpus of 462 journalistic maps used in news articles from five major news outlets published over three months. As a result, we created a design space comprised of eight dimensions that involved both properties describing the articles' aspects and the visual/interactive features of maps. We approach the second research question via semi-structured interviews with four data journalists who create data-driven articles daily. Through these interviews, we identified the most common design rationales made by editorial teams and potential gaps in current practices. We also collected the practitioners' feedback on our design space to externally validate it. With these results, we aim to provide researchers and journalists with empirical data to design and study journalistic maps.

Abstract:
Ensemble datasets are ever more prevalent in various scientific domains. In climate science, ensemble datasets are used to capture variability in projections under plausible future conditions including greenhouse and aerosol emissions. Each ensemble model run produces projections that are fundamentally similar yet meaningfully distinct. Understanding this variability among ensemble model runs and analyzing its magnitude and patterns is a vital task for climate scientists. In this paper, we present ClimateSOM, a visual analysis workflow that leverages a self-organizing map (SOM) and Large Language Models (LLMs) to support interactive exploration and interpretation of climate ensemble datasets. The workflow abstracts climate ensemble model runs-spatiotemporal time series-into a distribution over a 2D space that captures the variability among the ensemble model runs using a SOM. LLMs are integrated to assist in sensemaking of this SOM-defined 2D space, the basis for the visual analysis tasks. In all, ClimateSOM enables users to explore the variability among ensemble model runs, identify patterns, compare and cluster the ensemble model runs. To demonstrate the utility of ClimateSOM, we apply the workflow to an ensemble dataset of precipitation projections over California and the Northwestern United States. Furthermore, we conduct a short evaluation of our LLM integration, and conduct an expert review of the visual workflow and the insights from the case studies with six domain experts to evaluate our approach and its utility.

Abstract:
The detection and analysis of features in fluid flow are important tasks in fluid mechanics and flow visu alization. One recent class of methods to approach this problem is to first compute objective optimal reference frames, relative to which the input vector field becomes as steady as possible. However, existing methods either optimize locally over a fixed neighborhood, which might not match the extent of interesting features well, or perform global optimization, which is costly. We propose a novel objective method for the computation of optimal reference frames that automatically adapts to the flow field locally, without having to choose neighborhoods a priori. We enable adaptivity by formulating this problem as a moving least squares approximation, through which we determine a continuous field of reference frames. To incorporate fluid features into the computation of the reference frame field, we introduce the use of a scalar guidance field into the moving least squares approximation. The guidance field determines a curved manifold on which a regularly sampled input vector field becomes a set of irregularly spaced samples, which then forms the input to the moving least squares approximation. Although the guidance field can be any scalar field, by using a field that corresponds to flow features the resulting reference frame field will adapt accordingly. We show that using an FTLE field as the guidance field results in a reference frame field that adapts better to local features in the flow than prior wo rk. However, our moving least squares framework is formulated in a very general way, and therefore other types of guidance fields could be used in the future to adapt to local fluid features.

Abstract:
Visualizing and analyzing 3D unsteady flow fields is a very challenging task. We approach this problem by leveraging the mathematical foundations of 3D observer fields to explore and analyze 3D flows in reference frames that are more suitable to visual analysis than the input reference frame. We design novel interactive tools for determining, filtering, and combining reference frames for observer-aware 3D unsteady flow visualization. We represent the space of reference frame motions in a 3D spatial domain via a 6D parameter space, in which every observer is a time-dependent curve. Our framework supports operations in this 6D observer space by separately focusing on two 3D subspaces, for 3D translations, and 3D rotations, respectively. We show that this approach facilitates a variety of interactions with 3D flow fields. Building on the interactive selection of observers, we furthermore introduce novel techniques such as observer-aware streamline- and pathline-filtering as well as observer-aware isosurface animations of scalar fluid properties for the enhanced visualization and analysis of 3D unsteady flows. We discuss the theoretical underpinnings as well as practical implementation considerations of our approach, and demonstrate the benefits of its 6+1D observer-based methodology on several 3D unsteady flow datasets.

Abstract:
Chart corpora, which comprise data visualizations and their semantic labels, are crucial for advancing visualization research. However, the labels in most existing corpora are high-level (e.g., chart types), hindering their utility for broader applications in the era of AI. In this paper, we contribute VisAnatomy, a corpus containing 942 real-world SVG charts produced by over 50 tools, encompassing 40 chart types and featuring structural and stylistic design variations. Each chart is augmented with multi-level fine-grained labels on its semantic components, including each graphical element's type, role, and position, hierarchical groupings of elements, group layouts, and visual encodings. In total, VisAnatomy provides labels for more than 383k graphical elements. We demonstrate the richness of the semantic labels by comparing VisAnatomy with existing corpora. We illustrate its usefulness through four applications: semantic role inference for SVG elements, chart semantic decomposition, chart type classification, and content navigation for accessibility. Finally, we discuss research opportunities to further improve VisAnatomy.

Abstract:
Multimodal vision-language models (VLMs) continue to achieve ever-improving scores on chart understanding benchmarks. Yet, we find that this progress does not fully capture the breadth of visual reasoning capabilities essential for interpreting charts. We introduce EncQA, a novel benchmark informed by the visualization literature, designed to provide systematic coverage of visual encodings and analytic tasks that are crucial for chart understanding. EncQA provides 2,076 synthetic question-answer pairs, enabling balanced coverage of six visual encoding channels (position, length, area, color quantitative, color nominal, and shape) and eight tasks (find extrema, retrieve value, find anomaly, filter values, compute derived value exact, compute derived value relative, correlate values, and correlate values relative). Our evaluation of 9 state-of-the-art VLMs reveals that performance varies significantly across encodings within the same task, as well as across tasks. Contrary to expectations, we observe that performance does not improve with model size for many task-encoding pairs. Our results suggest that advancing chart understanding requires targeted strategies addressing specific visual reasoning gaps, rather than solely scaling up model or dataset size.

Abstract:
Recent work suggests that shape can encode quantitative data via a mapping between value and spatial frequency (SF). However, the set-size effect when perceiving multiple SF based items remains unclear. While automatic feature extraction has been found to be less affected by set size (number of items in a group), higher-level processes for making perceptual decisions tend to require increased cognitive demand. To investigate the set-size effect on comparing integrated SF based items, we used a risk-based scenario to assess discrimination performance. Participants were asked to discriminate between pairs of maps containing multiple SF glyphs, in which each glyph represents one of four discrete levels (none, low, medium, high), forming an aggregate “risk strength” per map. The set size was also adjusted across conditions, ranging from small (3 items) to large (7 items). Discrimination sensitivity is modeled with a logistic function and response time with a mixed-effect linear model. Results show that smaller set sizes and lower overall strength enable more precise discrimination, with faster response times for larger differences between maps. Incorporating set size and overall strength into the logistic model, we found that these variables both independently and jointly influence discrimination sensitivity. We suggest these results point towards capacity-limited processes rather than purely automatic ensemble coding. Our findings highlight the importance of set size and overall signal strength when presenting multiple SF glyphs in data visualization.

Abstract:
Analyzing literature involves tracking interactions between characters, locations, and themes. Visualization has the potential to facilitate the mapping and analysis of these complex relationships, but capturing structured information from unstructured story data remains a challenge. As large language models (LLMs) continue to advance, we see an opportunity to use their text processing and analysis capabilities to augment and reimagine existing storyline visualization techniques. Toward this goal, we introduce an LLM-driven data parsing pipeline that automatically extracts relevant narrative information from novels and scripts. We then apply this pipeline to create Story Ribbons, an interactive visualization system that helps novice and expert literary analysts explore detailed character and theme trajectories at multiple narrative levels. Through pipeline evaluations and user studies with Story Ribbons on 36 literary works, we demonstrate the potential of LLMs to streamline narrative visualization creation and reveal new insights about familiar stories. We also describe current limitations of AI-based systems, and interaction motifs designed to address these issues.

Abstract:
When a reader encounters a word in English, they split the word into smaller orthographic units in the process of recognizing its meaning. For example, “rough”, when split according to phonemes, is decomposed as r-ou-gh (not as r-o-ugh or r-ough), where each group of letters corresponds to a sound. Since there are many ways to segment a group of letters, this constitutes a computational operation that has to be solved by the reading brain, many times per minute, in order to achieve the recognition of words in text necessary for reading. In English, the irregular relationships between groups of letters and sounds, and the wide variety of possible groupings make this operation harder than in more regular languages such as Italian. If this segmentation takes a significant amount of time in the process of recognizing a word, it is conceivable that providing segmentation information in the text itself could help the reading process by reducing its computational cost. In this paper we explore whether and how different visual interventions from the visualization literature could communicate segmentation information for reading and word recognition. We ran a series of pre-registered lexical decision experiments with 192 participants that tested five main types of visual segmentations: outlines, spacing, connections, underlines and color. The evidence indicates that, even with a moderate amount of training, these visual interventions always slow down word identification, but each to a different extent (between 32.7ms—color technique—and 70.7ms—connection technique). These findings are important because they indicate that, at least for typical adult readers with a moderate amount of specific training in these visual interventions, accelerating the lexical decision task is unlikely. Importantly, the results also offer an empirical measurement of the cost of a common set of visual manipulations of text, which can be useful for practitioners seeking to visualize alongside or within text without impacting reading performance. Finally, the interaction between typographically encoded information and visual variables presented unique patterns that deviate from existing theories, suggesting new directions for future inquiry.

Abstract:
We present VizGenie, a self-improving, agentic framework that advances scientific visualization through large language model (LLM) by orchestrating of a collection of domain-specific and dynamically generated modules. Users initially access core functionalities—such as threshold-based filtering, slice extraction, and statistical analysis—through pre-existing tools. For tasks beyond this baseline, VizGenie autonomously employs LLMs to generate new visualization scripts (e.g., VTK Python code), expanding its capabilities on-demand. Each generated script undergoes automated backend validation and is seamlessly integrated upon successful testing, continuously enhancing the system's adaptability and robustness. A distinctive feature of VizGenie is its intuitive natural language interface, allowing users to issue high-level feature-based queries (e.g., “visualize the skull” or “highlight tissue boundaries”). The system leverages image-based analysis and visual question answering (VQA) via fine-tuned vision models to interpret these queries precisely, bridging domain expertise and technical implementation. Additionally, users can interactively query generated visualizations through VQA, facilitating deeper exploration. Reliability and reproducibility are further strengthened by Retrieval-Augmented Generation (RAG), providing context-driven responses while maintaining comprehensive provenance records. Evaluations on complex volumetric datasets demonstrate significant reductions in cognitive overhead for iterative visualization tasks. By integrating curated domain-specific tools with LLM-driven flexibility, VizGenie not only accelerates insight generation but also establishes a sustainable, continuously evolving visualization practice. The resulting platform dynamically learns from user interactions, consistently enhancing support for feature-centric exploration and reproducible research in scientific visualization.

Abstract:
Designing and building visual analytics (VA) systems is a complex, iterative process that requires the seamless integration of data processing, analytics capabilities, and visualization techniques. While prior research has extensively examined the social and collaborative aspects of VA system authoring, the practical challenges of developing these systems remain underexplored. As a result, despite the growing number of VA systems, there are only a few structured knowledge bases to guide their design and development. To tackle this gap, we propose VA-Blueprint, a methodology and knowledge base that systematically reviews and categorizes the fundamental building blocks of urban VA systems, a domain particularly rich and representative due to its intricate data and unique problem sets. Applying this methodology to an initial set of 20 systems, we identify and organize their core components into a multi-level structure, forming an initial knowledge base with a structured blueprint for VA system development. To scale this effort, we leverage a large language model to automate the extraction of these components for other 81 papers (completing a corpus of 101 papers), assessing its effectiveness in scaling knowledge base construction. We evaluate our method through interviews with experts and a quantitative analysis of annotation metrics. Our contributions provide a deeper understanding of VA systems' composition and establish a practical foundation to support more structured, reproducible, and efficient system development. VA-Blueprint is available at urbantk.org/va-blueprint.

Abstract:
In-vitro fertilization (IVF) has become standard practice to address infertility, which affects more than one in ten couples in the US. However, current protocols yield relatively low success rates of about 20% per treatment cycle. A critical but complex and time-consuming step is the grading and selection of embryos for implantation. Although incubators with time-lapse microscopy have enabled computational analysis of embryo development, existing automated approaches either require extensive manual annotations or use opaque deep learning models that are hard for clinicians to validate and trust. We present EmbryoProfiler, a visual analytics system collaboratively developed with embryologists, biologists, and machine learning researchers to support clinicians in visually assessing embryo viability from time-lapse microscopy imagery. Our system incorporates a deep learning pipeline that automatically annotates microscopy images and extracts clinically interpretable features relevant for embryo grading. Our contributions include: (1) a semi-automatic, visualization-based workflow that guides clinicians through fertilization assessment, developmental timing evaluation, morphological inspection, and comparative analysis of embryos; (2) innovative interactive visualizations, such as cell-shape plots, designed to facilitate efficient analysis of morphological and developmental characteristics; and (3) an integrated, explainable machine learning classifier offering transparent, clinically-informed embryo viability scoring to predict live birth outcomes. Quantitative evaluation of our classifier and qualitative case studies conducted with practitioners demonstrate that EmbryoProfiler enables clinicians to make better-informed embryo selection decisions, potentially leading to improved clinical outcomes in IVF treatments.

Abstract:
Effective data visualizations enhance perception, support cognitive processing, and facilitate informed decision-making by aligning with human perceptual strengths. Conversely, poorly designed visualizations can impede comprehension, introduce interpretive bias, and diminish the perceived credibility of the conveyed message. This paper investigates the extent to which visual embellishments influence perceived message credibility in data visualizations. We conducted two crowdsourced experiments to examine both holistic and component-level effects of embellishment. In the first experiment, participants evaluated the relative credibility of plain bar charts versus two embellished variants-cartoon-style and image-style-across topics. Participants provided both comparative judgments and qualitative feedback. In the second experiment, we systematically isolated the influence of specific design elements-color, font, and bar style-on credibility perceptions through controlled variations. Our findings reveal that the impact of embellishments on perceived message credibility is complex and context-dependent. While certain embellishments, such as the use of color and image style bars, enhanced credibility, others-most notably hand-drawn fonts and cartoon-style bars-significantly undermined it. By operationalizing trust through the lens of message credibility, this work offers empirical insight into the design factors that shape viewers' perceptions. We conclude by proposing actionable design guidelines to support the creation of visualizations that are effective for communication and credible.

Abstract:
We contribute an autoethnographic reflection on the complexity of defining and measuring visualization literacy (i.e., the ability to interpret and construct visualizations) to expose our tacit thoughts that often exist in-between polished works and remain unreported in individual research papers. Our work is inspired by the growing number of empirical studies in visualization research that rely on visualization literacy as a basis for developing effective data representations or educational interventions. Researchers have already made various efforts to assess this construct, yet it is often hard to pinpoint either what we want to measure or what we are effectively measuring. In this autoethnography, we gather insights from 14 internal interviews with researchers who are users or designers of visualization literacy tests. We aim to identify what makes visualization literacy assessment a “wicked” problem. We further reflect on the fluidity of visualization literacy and discuss how this property may lead to misalignment between what the construct is and how measurements of it are used or designed. We also examine potential threats to measurement validity from conceptual, operational, and methodological perspectives. Based on our experiences and reflections, we propose several calls to action aimed at tackling the wicked problem of visualization literacy measurement, such as by broadening test scopes and modalities, improving test ecological validity, making it easier to use tests, seeking interdisciplinary collaboration, and drawing from continued dialogue on visualization literacy to expect and be more comfortable with its fluidity.

Abstract:
This paper evaluates the visualization literacy of modern Large Language Models (LLMs) and introduces a novel prompting technique called Charts-of-Thought. We tested three state-of-the-art LLMs (Claude-3.7-sonnet, GPT-4.5-preview, and Gemini-2.0-pro) on the Visualization Literacy Assessment Test (VLAT) using standard prompts and our structured approach. The Charts-of-Thought method guides LLMs through a systematic data extraction, verification, and analysis process before answering visualization questions. Our results show Claude-3.7-sonnet achieved a score of 50.17 using this method, far exceeding the human baseline of 28.82. This approach improved performance across all models, with score increases of 21.8% for GPT-4.5, 9.4% for Gemini-2.0, and 13.5% for Claude-3.7 compared to standard prompting. The performance gains were consistent across original and modified VLAT charts, with Claude correctly answering 100% of questions for several chart types that previously challenged LLMs. Our study reveals that modern multimodal LLMs can surpass human performance on visualization literacy tasks when given the proper analytical framework. These findings establish a new benchmark for LLM visualization literacy and demonstrate the importance of structured prompting strategies for complex visual interpretation tasks. Beyond improving LLM visualization literacy, Charts-of-Thought could also enhance the accessibility of visualizations, potentially benefiting individuals with visual impairments or lower visualization literacy.

Abstract:
With the advancement of haptic interfaces, recent studies have focused on enabling detailed haptic experiences in virtual reality (VR), such as fluid-haptic interaction. However, rendering forces from fluid contact often causes a high-cost computation. Given that motion-induced fluid feedback is crucial to the overall experience, we focus on hand-perceivable forces to enhance underwater haptic sensation by achieving high-fidelity rendering while considering human perceptual capabilities. We present a new multimodal (tactile and kinesthetic) haptic rendering pipeline. Here, we employ drag and added mass forces by dynamically adapting to the user’s hand movement and posture with pneumatic-based haptic gloves. We defined decaying and damping effects to indicate fluid properties caused by inertia and confirmed their significant perceptual impacts compared to using only physics-based equations in a perception study. By modulating pressure variations, we reproduced fluid smoothness via exponential tactile deflation and light fluid mass via linear kinesthetic feedback. Our pipeline enabled richer and more immersive VR underwater experiences by accounting for precise hand regions and motion diversity.

Abstract:
Data-parallel ray tracing is a crucial technique for rendering large-scale scenes that exceed the memory capacity of a single compute node. It partitions scene data across multiple nodes and accesses remote data through inter-node communication. However, the resulting communication overhead remains a significant bottleneck for practical performance. Existing approaches mitigate this bottleneck by enhancing data locality through dynamic scheduling during rendering, typically employing spatial partitioning to enable access prediction. Although effective in some scenarios, these methods incur significant redundancy in base geometry when applied to large-scale instanced scenes. In this paper, we introduce the first object-space-based dynamic scheduling algorithm, which uses object groups as the scheduling units to eliminate redundant storage of base data in instanced scenes. Additionally, we propose two data access frequency prediction methods to guide asynchronous data prefetching, enhancing rendering efficiency. Compared to the state-of-the-art method, our approach achieves an average rendering speedup of 77.6%, with a maximum improvement of up to 146.1%, while incurring only a 5% increase in scene memory consumption.

Abstract:
Depth completion which aims at predicting dense depth maps from sparse depth measurements, plays a crucial role in many computer graphics and computer vision applications. Previous supervised learning based approaches have demonstrated overwhelming success in this task, while unsupervised high-precision depth completion without relying on the ground-truth data still remains challenging. One main drawback of most previous unsupervised solutions comes from the ignorance of 3D structural information, which often leads to inaccurate spatial propagation and mixed-depth problems. To alleviate the above challenges, this paper explores the utilization of 3D perceptual features and multi-view geometry consistency to devise a high-precision self-supervised depth completion method. Our key contribution is a 3D perceptual spatial propagation constructed with a point cloud representation and an attention weighting mechanism, to capture more reasonable and favorable neighbors during the depth propagation process. Based on the 3D perceptual spatial propagation, we also introduce multi-view geometric constraints between adjacent views to guide the optimization of the whole depth completion model, which achieves geometry consistent depth completion in a self-supervised manner. Extensive experiments on benchmark datasets of NYU-Depth-v2, VOID and KITTI Depth Completion demonstrate that the proposed model achieves the state-of-the-art depth completion performance compared with other unsupervised methods, and even competitive performance compared with previous supervised methods.

Abstract:
Creating interactive scenes often involves complex programming tasks. Although large language models (LLMs) like ChatGPT can generate code from natural language, their output is often error-prone, particularly when scripting interactions among multiple elements. The linear conversational structure limits the editing of individual elements, and the lack of graphical and precise control complicates visual integration. To address these issues, we integrate a context-aware modularization technique that processes textual descriptions for individual elements through separate LLM modules, with a central module managing interactions among elements. It defines a top-down structure to manage interactions, ensuring clear update logic and facilitating efficient collaboration while allowing for independent updates for each element. We design a graphical user interface, MoGraphGPT, which combines modular LLMs with enhanced graphical control to generate codes for 2D interactive scenes. It enables direct integration of graphical information and offers quick, precise control through automatically generated sliders. A comparative study with Cursor Composer shows MoGraphGPT significantly improves easiness, controllability, and performance in creating 2D interactive scenes with multiple visual elements in a coding-free manner. An ablation study validates the effectiveness of modularization, and an open-ended study demonstrates the usability and expressiveness of MoGraphGPT.

Abstract:
We propose a practical spectral image rendering method for fluorescent objects using a renderer that does not natively support wavelength-shifting transport. For scenes composed of matte fluorescent and non-fluorescent surfaces, the illumination incident on a target fluorescent object is classified into three types: (1) direct illumination from a light source, (2) indirect illumination reflected from other reflective objects, and (3) luminescent illumination emitted from other fluorescent objects. We express the radiance observed on a fluorescent surface as a linear combination of five terms: reflection and fluorescence under direct illumination, reflection and fluorescence under indirect illumination, and reflection induced by luminescent illumination emitted by other fluorescent surfaces. Assuming isotropic emission and single-bounce fluorescence, the fluorescent shading terms reuse the diffuse-reflection shading produced by the renderer, while measured Donaldson matrices provide the wavelength conversion. Mitsuba is used as the underlying conventional rendering system. Experiments conducted with a physically built Cornell Box validate the proposed method through comparisons with direct measurements and an existing fluorescence-capable renderer. Finally, we demonstrate an extension to non-planar fluorescent objects via sparse emitter discretization.

Abstract:
We propose a novel method to generate a small set of ruled surfaces that do not collide with the input shape for linear hot-wire rough machining. Central to our technique is a new observation: ruled surfaces constructed by vertical extrusion from planar smooth curves that approach the input shape’s outer contour lines without collisions can effectively remove material during rough machining. Accordingly, we develop an iterative algorithm that alternates in each iteration between computing a viewpoint to determine an outer contour line and optimizing a smooth curve to approximate that contour line under the collision-free constraint. Specifically, a view selection approach based on a genetic algorithm is used to optimize the viewpoint for removing materials as much as possible, and an adaptive fitting algorithm is presented to find the constrained curves. The feasibility and practicability of our method are demonstrated through 10 physical examples. Compared with manual designs, our method obtains lower errors with the same number of cuts.

Abstract:
This paper presents MonoRelief V2, an end-to-end model designed for directly recovering 2.5D reliefs from single images under complex material and illumination variations. In contrast to its predecessor, MonoRelief V1 (Gao et al. 2025), which was solely trained on synthetic data, MonoRelief V2 incorporates real data to achieve improved robustness, accuracy and efficiency. To overcome the challenge of acquiring large-scale real-world dataset, we generate approximately 15,000 pseudo-real images using a text-to-image generative model, and derive corresponding depth pseudo-labels through fusion of depth and normal predictions. Furthermore, we construct a small-scale real-world dataset (800 samples) via multi-view reconstruction and detail refinement. MonoRelief V2 is then progressively trained on the pseudo-real and real-world datasets. Comprehensive experiments demonstrate its state-of-the-art performance both in depth and normal predictions, highlighting its strong potential for a range of downstream applications.

Abstract:
Text-guided texture generation has been rapidly developed with the proliferation of generative artificial intelligence for creating three-dimensional textured objects. However, existing text-guided texture generation methods often suffer from artifacts such as inconsistent visual appearance across different views, the Janus problems and seams in texture maps. To address these issues, we propose a novel text-guided texture generation method, named WonderTex. It achieves the generation of high-quality, view-consistent, and seamless texture maps through a two-stage pipeline. Specifically, we fine-tune a Stable Diffusion model using a large dataset to obtain a multi-view image diffusion model capable of generating a 4-view grid. This model serves as the foundation for producing four consistent views and establishing the base texture in the first stage. Subsequently, an automatic view selection and inpainting strategy is employed to effectively fill and refine the texture maps in the second stage. Extensive experiments have shown that our method is effective and robust, capable of generating high-quality textures with various meshes and prompts, outperforming baseline methods in terms of texture details, view consistency, and other metrics.

Abstract:
Estimating the surface of translucent objects from photometric data poses significant challenges due to complex internal light scattering. We introduce a novel method that computes a depth map from single-viewpoint photographs of a material sample, captured under multiple illuminations. Our approach leverages inverse rendering to derive a volumetric representation, including density, albedo, and phase function, from which a surface mesh is reconstructed. Beyond validation with synthetic and 3D-printed physical models, we illustrate our technique’s power by successfully applying it to the digitization of fabrics, a notoriously difficult material due to its intricate translucent structure. This work advances the state-of-the-art texture stack acquisition via enhanced surface reconstruction.

Abstract:
In modern computer-aided design (CAD), Boundary Representation (B-rep) is a widely used geometric modeling technique in industrial design and manufacturing. However, existing B-rep generation methods, which rely on tree-based hierarchies to represent and generate B-reps, fail to fully exploit the inherent graph structure of B-reps, resulting in suboptimal efficiency and model quality. To address this issue, we propose BRep-GD, a graph diffusion-based model specifically designed for B-rep generation. Unlike prior methods, BRep-GD treats B-reps as graphs, where nodes represent face elements, and edges represent boundary and vertex elements. By utilizing a continuous topological graph data structure, BRep-GD overcomes the challenges associated with directly applying graph diffusion models to B-rep generation. Specifically, BRep-GD introduces a graph diffusion method tailored to the features of CAD data, generating faces and edges sequentially. During edge generation, continuous topology decoupling is employed to avoid the need for global attention calculations, reducing computational complexity while ensuring geometric consistency and high-quality results. Experimental results demonstrate that BRep-GD outperforms existing state-of-the-art methods in both unconditional and class-conditional generation tasks, particularly in generating watertight solids and handling complex geometries. It significantly reduces isolated or inconsistent geometric components and improves generation efficiency.

Abstract:
Portrait relighting shows great potential in photography, film, and AR by simulating diverse lighting effects. Existing state-of-the-art methods often rely on expensive paired OLAT or synthetic data, which limits scalability. Moreover, accurately modeling the interaction between physics-guided rendering, neural rendering, and real-world remains challenging. To address these issues, we propose a novel multi-stage self-supervised relighting framework. It progressively refines intrinsic scene properties via a simple-to-complex training strategy, removing the need for expensive paired data while adapting to various lighting conditions. One core design introduces a novel pre-training method approach using diverse shading-based masking for self-reconstruction, which improves the model’s perception of complex lighting variations. Furthermore, we introduce two perceptual modules that leverage the linear superposition of light to narrow the gap between physics-guided and neural rendering, and better align relit results with real-world observations. Extensive experiments demonstrate that our unified framework achieves new state-of-the-art performance in portrait relighting, surpassing recent methods in photorealism, synthesis quality, and identity preservation. It provides a practical paradigm for high-fidelity relighting under diverse lighting.

Abstract:
Interest in, and the need for, remote learning tools and techniques, as well as the adoption and deployment of digital technologies for tutoring and training, continues to rise. Concurrently, over the past decade, the use of Virtual Reality (VR) applications as a supplementary tool in the learning process has increased significantly. Therefore, VR simulation applications are expected to play a key role in people’s education and training in the coming years. In this study, we designed and developed a VR training application called ViRtus –a VR-based system following constructivist and serious game human-computer interaction principles for VR-aided training– and applied it to a real-world scenario, i.e., the construction of an industrial electrical control panel. To demonstrate the potential and sustainability of the VR training system as an alternative to traditional apprenticeship training procedures, we conducted an experimental study comparing the real-world outcomes of three groups trained using a conventional method, a VR-only training, and a hybrid approach combining both. Based on qualitative assessments and the statistical analysis of practical experiments, the collected data and observations indicate that the hybrid VR constructivist training strategy –supplementing conventional trainer instructions with the VR application– can enhance training effectiveness and promote a more productive and sustainable workplace, for both trainees and trainers, in high-risk industrial tasks.

Abstract:
Modeling and simulating realistic insect flight pose unique challenges due to the complex interaction between multi-degree-of-freedom wing kinematics and highly precise aerodynamic forces. To solve this challenge, this article presents a bidirectional kinematics–aerodynamics coupled simulation framework for miniature insect flight. Our approach first models the kinematics of flying insects by parameterizing natural wingbeat cycles based on available real-world datasets. Subsequently, we compute aerodynamic forces utilizing an improved semi-empirical model, which extends from quasi-steady formulation by incorporating critical unsteady force components. To achieve closed-loop control for both kinematics and aerodynamics, we employ deep reinforcement learning to train a virtual insect to adaptively adjust flapping strategies in response to dynamic flight states. Finally, an integrated controller enables the simulated insect to autonomously regulate the wing motion and perform complex tasks such as visual obstacle avoidance. Extensive experiments and comparisons demonstrate that our framework can effectively generate physically plausible and autonomous insect flight across a variety of scenarios.

Abstract:
Automatic real image editing offers unprecedented freedom to modify the appearance of the image or to edit a few objects through natural language. Recent scalable model families such as diffusion models have showcased remarkable proficiency in editing highly realistic images due to the introduction of vast amounts of training data and large pretrained language models. However, these large diffusion models require iterative evaluation that would significantly hinder the pace of image editing. Moreover, the pioneering work in this field necessitates the learning of a unique textual token that corresponds to each input image, or a group of images containing the same object, leading to the generation of redundant and fragmented models. Given the aforementioned problems, we suggest a novel Large pretrained models Assistant text-guided image Editing adversarial Network (LAE-Net) in this paper. More concretely, we introduce a deep semantic editing network to globally transfer text information among different isolated editing blocks, which would extract features from the source image to differentiate text-required areas from text-irrelevant ones. Furthermore, based on idea that the multi-modal CLIP model, leveraging vision-language alignment, captures comprehensive global semantic cues, whereas the vision-centric DINO model specializes in delivering intricate, fine-grained pixel-level details, the powerful discriminator of LAE-Net is designed by harnessing the visual embeddings derived from both the CLIP and DINO models separately to boost the visual discriminant capability and facilitate training a strong generator for conditioning image generation. Comprehensive experimental evaluations show that our LAE-Net not only delivers outstanding performance but also surpasses several cutting-edge models.

Abstract:
The integrity of data visualizations is increasingly threatened by image editing techniques that enable subtle yet deceptive tampering. Through a formative study, we define this challenge and categorize tampering techniques into two primary types: data manipulation and visual encoding manipulation. To address this, we present VizDefender, a framework for tampering detection and analysis. The framework integrates two core components: 1) a semi-fragile watermark module that protects the visualization by embedding a location map to images, which allows for the precise localization of tampered regions while preserving visual quality, and 2) an intent analysis module that leverages Multimodal Large Language Models (MLLMs) to interpret manipulation, inferring the attacker's intent and misleading effects. Extensive evaluations and user studies demonstrate the effectiveness of our methods.

Abstract:
Density plots effectively summarize large numbers of points, which would otherwise lead to severe overplotting in, for example, a scatter plot. However, when applied to line-based datasets, such as trajectories or time series, density plots alone are insufficient, as they disrupt path continuity, obscuring smooth trends and rare anomalies. We propose a bin-based illumination model that decouples structure from density to enhance flow and reveal sparse outliers while preserving the original colormap. We introduce a bin-based outlierness metric to rank trajectories. Guided by this ranking, we construct a structural normal map and apply locally-adaptive lighting in the luminance channel to highlight chosen patterns-from dominant trends to atypical paths-with acceptable color distortion. Our interactive method enables analysts to prioritize main trends, focus on outliers, or strike a balance between the two. We demonstrate our method on several real-world datasets, showing it reveals details missed by simpler alternatives, achieves significantly lower CIEDE2000 color distortion than standard shading, and supports interactive updates for up to 10,000 lines.

Abstract:
Unstructured volumetric meshes serve as fundamental data representations in various scientific simulations and analyses. They play a crucial role in representing complex computational domains and are essential for important numerical techniques, such as finite element analysis. Whenever such a mesh is read from a file, streamed in-situ, or generated by algorithms, scientific visualization libraries rely on calculating the external surface of a geometry, named “external facelist”, to produce a polygonal mesh for rendering. Consequently, external facelist calculation has become one of the most widely used algorithms in the scientific visualization domain, necessitating optimal performance. In this paper, we explore relevant work on external facelist calculation algorithms in two common visualization libraries, VTK and Viskores, assess their performance and memory constraints, and introduce a novel memory-aware external facelist calculation algorithm employing an atomic hash counting approach. This algorithm fully leverages Viskores' data-parallel primitive operations, facilitating its execution across diverse many-core architectures. Our algorithm features the lowest memory footprint on the GPU and the second-lowest on the CPU among all evaluated methods, and it also delivers the fastest performance on both CPU and GPU. It has been made available under an open-source license in the VTK and Viskores visualization systems.

Abstract:
Visualizing the large-scale datasets output by HPC resources presents a difficult challenge, as the memory and compute power required become prohibitively expensive for end user systems. Novel view synthesis techniques can address this by producing a small, interactive model of the data, requiring only a set of training images to learn from. While these models allow accessible visualization of large data and complex scenes, they do not provide the interactions needed for scientific volumes, as they do not support interactive selection of transfer functions and lighting parameters. To address this, we introduce Volume Encoding Gaussians (VEG), a 3D Gaussian-based representation for volume visualization that supports arbitrary color and opacity mappings. Unlike prior 3D Gaussian Splatting (3DGS) methods that store color and opacity for each Gaussian, VEG decouple the visual appearance from the data representation by encoding only scalar values, enabling transfer function-agnostic rendering of 3DGS models. To ensure complete scalar field coverage, we introduce an opacity-guided training strategy, using differentiable rendering with multiple transfer functions to optimize our data representation. This allows VEG to preserve fine features across a dataset's full scalar range while remaining independent of any specific transfer function. Across a diverse set of volume datasets, we demonstrate that our method outperforms the state-of-the-art on transfer functions unseen during training, while requiring a fraction of the memory and training time.

Abstract:
Projection mapping (PM) enables augmented reality (AR) experiences without requiring users to wear head-mounted displays and supports multi-user interaction. It is regarded as a promising technology for a variety of applications in which users interact with content superimposed onto augmented objects in tabletop workspaces, including remote collaboration, healthcare, industrial design, urban planning, artwork creation, and office work. However, conventional PM systems often suffer from projection shadows when users occlude the light path. Prior approaches employing multiple distributed projectors can compensate for occlusion, but suffer from latency due to computational processing, degrading the user experience. In this research, we introduce a synthetic-aperture PM system that uses a significantly larger number of projectors, arranged densely in the environment, to achieve delay-free, shadowless projection for tabletop workspaces without requiring computational compensation. To address spatial resolution degradation caused by subpixel misalignment among overlaid projections, we develop and validate an offline blur compensation method whose computation time remains independent of the number of projectors. Furthermore, we demonstrate that our shadowless PM plays a critical role in achieving a fundamental goal of PM: altering material properties without evoking projection-like impression. Specifically, we define this perceptual impression as “sense of projection (SoP)” and establish a PM design framework to minimize the SoP based on user studies.

Abstract:
This paper presents an approach to alleviating the server-client communication bottleneck in collocated multi-user virtual reality (VR). Instead of transmitting unique packets to each client, the system multicasts the same packets to all users, combining packetization, visibility-aware prioritization, and lightweight acknowledgments to ensure efficiency and fairness. To enable rapid scene completion, the environment is partitioned into independent fixed-size packets that can be decoded immediately upon arrival, and transmission order is guided by precomputed visibility footprints. Scalability is achieved through repeated multicast until compact bitmap acknowledgments confirm reception, keeping communication cost mostly independent of the number of clients. A controlled study with 23 participants in groups of 7 and 16 showed that this method reconstructs environments several times faster and with fewer missing parts than a conventional unicast TCP-based approach. These findings demonstrate that multicast transmission over commodity Wi-Fi can support large collocated VR groups.

Abstract:
Avatars play a central role in virtual and augmented reality (VR/AR), yet little is known about how they are represented in research. To address this gap, we collected and analyzed 14 440 avatar images from 4 659 publications. Every image was hand-labeled for gender, ethnicity, body representation, and visual style, and linked to publication metadata such as keywords, affiliation, and year. Our large-scale analysis yields 14 findings: for example, the dominance of realistic white male avatars in VR studies, the use of lower-fidelity and more gender-diverse bodies in AR, the post-2021 rise of stylized designs and diversity labels, and the mismatch between the amount of female keywords and their avatar occurrences. Based on our findings, we offer practical implications for avatar designers and researchers, such as adopting balanced starter libraries, employing bias dashboards, and using simple representation checklists. These steps can help future VR and AR platforms reduce bias and improve representations in avatar systems.

Abstract:
As 3D Gaussian Splatting (3DGS) emerges as a leading approach for novel view synthesis and scene reconstruction, its potential in digital asset creation has gained significant attention. An increasing number of asset libraries based on GS are being established. However, generating physics-based dynamic assets remains a time-consuming and expertise-intensive task, especially for non-experts. In this paper, we propose LIVE-GS, a highly realistic Virtual Reality (VR) system powered by Large Language Models (LLMs), which enables rapid creation of dynamic Gaussian assets and real-time VR interactions. To inform our system design, we conducted interviews to examine challenges faced by current GS-based VR systems and the specific demands of users. Based on these insights, we employed GPT-4o to analyze key physical properties of objects that significantly impact user interactions, ensuring physics-based interactions in VR align with real-world phenomena. A key innovation of LIVE-GS is its ability to predict reasonable parameters in just 10 seconds from static Gaussian assets while maintaining high-quality VR interactions. To validate our approach, we invited participants experienced in physical simulation to manually adjust physical parameters, providing a baseline for comparison in both asset quality and authoring efficiency. We also conducted a comprehensive user study to evaluate system usability and user satisfaction. Experimental results demonstrate that LIVE-GS, leveraging LLMs' scene understanding capabilities, can achieve efficient physical scene creation and natural interactions without requiring manual design or annotation.

Abstract:
With the increasing emphasis on real-world applications of Extended Reality (XR), the field faces growing challenges in creating meaningful, user-centred, and ethical experiences. Understanding how human values have been considered in prior research is essential for ensuring that future XR experiences are designed with a deep awareness of human needs, behaviours, and cultural contexts. This study presents a systematic review of human values in the extant XR literature. It aims to address which values have been explored, how they have been studied, and why they have been considered in XR contexts. We report findings on authors' research motivations, approaches to studying values, application domains, forms of technology, target groups and associated values, as well as identified directions for future research. Based on these findings, we propose a structured analysis that synthesises key trends and gaps, providing a foundation for future research to advance more inclusive, ethically grounded, and context-aware XR design.

Abstract:
Cybersickness has been a hindrance to the widespread adoption of virtual reality. As cybersickness is dynamic, with its manifestation changing constantly, predicting and mitigating it in real time is key. However, previous research has utilized predictive models trained mostly with datasets whose sickness levels were measured and labeled only after experiencing the content. In addition, many datasets rely on physiological signals as input, which makes them difficult to apply in actual VR usage. This makes such timely predictions unreliable or practically infeasible. We have created a publicly available dataset where the ground truth sickness levels were densely marked every 0.5 seconds and adjusted/updated on-demand using the FMS, and purposely excluded the difficult-to-collect physiological data. We demonstrate that predictive models trained with such a dataset, comprising just the content motion profile and FMS data, can still produce comparably reliable sickness prediction, and more so, when user-specific parameters (e.g., age, gender) are added. The created dataset is available at https://github.com/pvtryan1024/2025_Cybersickness_dataset.

Abstract:
In cloud virtual reality (VR), delivering high perceptual visual quality under constrained wireless bandwidth remains a pivotal challenge. Foveated video encoding (FVE) copes with this by human vision-driven quality allocation, reducing bandwidth demand while preserving high visual fidelity for immersive VR streaming. However, existing FVE methods are content-independent, relying exclusively on gaze position and predefined quality degradation profiles. Thus, they fall short in modeling the intricate mechanisms of human attention. This leads to suboptimal quality distribution for visual saliency stimuli, causing perceptible artifacts and degraded perceptual quality. In this paper, we introduce a novel AI-driven streaming framework that incorporates visual saliency cues, defined as the innate capacity of scene elements to attract attention, into the video encoding pipeline. To meet the stringent low-latency demands of immersive VR, we propose a lightweight deep neural network for saliency inference, reducing computational complexity (FLOPs) by 48× compared to prior models while maintaining comparable accuracy. We integrate our pipeline into an open-source cloud VR gaming platform and conduct comprehensive experiments. Evaluation results demonstrate that our approach enhances perceptual visual quality by 22.98% compared to SoTA systems. Our IRB-approved user study shows that the saliency-guided FVE pipeline achieves superior visual quality and spatial smoothness while significantly reducing noticeable artifacts introduced by gaze-exclusive FVE. The project source code is available at https://github.com/WuZemyp/SaliencyFov.

Abstract:
Traditional target-selection-based teleportation depends on the intersection of a (curved) ray with the scene's geometry, which limits navigation to points on the ground, restricting users' navigational freedom. While previous techniques exist that permit mid-air target selection, they are not optimal for transitioning between air and ground navigation, leading to inaccurate or lengthy interaction sequences. In this paper, we introduce SATOR, a new 3D teleportation technique designed to enable efficient and accurate navigation to both ground and mid-air targets by combining and enhancing previous approaches. Informed by the literature, we implemented four different parametrizations of our technique and compared them to a previously published technique that also enables both ground and mid-air target selection. Our user study with 30 participants indicates that SATOR is more efficient, accurate, and easier to use than the baseline. As a result, SATOR effectively helps users get an overview of the environment, observe features at different heights, or maneuver quickly around larger obstacles.

Abstract:
Virtual reality is increasingly adopted in education for its potential to create immersive, interactive learning experiences. Among various instructional approaches, gamification has emerged as a promising strategy to enhance learner engagement and motivation. While VR naturally provides a platform for game-based learning, current applications often overlook the underlying motivational mechanisms that influence learner behavior in these environments. Building on Self-Determination Theory, we examined how gamification and motivational framing influence VR learning. We designed a VR system featuring both gamified and non-gamified versions of two learning tasks—one culturally expressive (batik-based task) and one technically focused (code-based task)—to reflect different motivational framings. Our findings show that gamification selectively influenced intrinsic motivation components and played a dominant role in shaping learner satisfaction. A trend-level interaction suggested that combining extrinsic framing with gamification may increase satisfaction. Gamification improved overall experience and engagement, even though it elevated perceived workload. These results highlight the importance of aligning gamification strategies with learner motivation and underscore the need for more dynamic approaches to capture motivational processes in VR, offering design insights for more effective and psychologically attuned VR learning environments.

Abstract:
Generative content is increasingly applied in VR to provide immersive experiences, yet maintaining high generation quality remains challenging for audiovisual effects. Particularly in dynamic fluid phenomena, achieving realism and presence requires adherence to physical laws. To accomplish this objective, this work proposes an audiovisual synergy fluid animation generation framework, which enhances immersion by improving motion texture fidelity and audiovisual consistency. It comprises Detail-Enhanced Texture generator (DET) and Physics-Guided Audio generator (PGA). DET integrates Global-Local Physics guidance (GLP) and Temporal Texture Modeling (TTM) to produce video textures, explicitly optimizing dynamic details by leveraging local motion cues and assigned cumulative differences. PGA incorporates Visual Semantic Augmenter (VSA) and Rhythm Semantic Adapter (RSA) to synchronize audio by fusing static visual semantics with dynamic motion semantics to improve temporal coherence. By integrating DET and PGA, this framework strengthens audiovisual immersion in VR natural dynamic scenes from both visual and auditory perspectives. Quantitative and qualitative evaluations demonstrate that our approach surpasses most existing methods in terms of texture realism and audiovisual synchronization, offering new insights for advancing immersive experiences in dynamic VR phenomena.

Abstract:
A central promise of virtual reality (VR) games is the increased control players have over their character through pose and body language. However, many non-player character (NPC) systems fail to respond convincingly to these poses, user intent, and situational context, limiting immersion. We present Look2React, an interaction system that captures what NPCs see, using a vision-based reasoning model to select pose and text responses. Look2React endows NPCs with the ability to react dynamically and appropriately to player interactions. Through a gaze and proximity-based detection system inspired by stealth games, we trigger our system intuitively and only when intended, while also reducing resource costs. We invited 20 participants to play two versions of an RPG game: one with NPCs based on contemporary games and the other with Look2React NPCs. Our results demonstrate that Look2React increases engagement, leading to more frequent and repeated interactions with NPCs. Participants reported more satisfactory play sessions, significantly increased feelings of social presence, and felt that the dynamic reactions gave the NPCs more depth and personality — ultimately making them feel more human.

Abstract:
Understanding quantum mechanics is inherently challenging due to its counterintuitive principles. Quantum Intuition XR is an interactive, extended reality (XR) experience designed to make quantum concepts tangible. Our system visualizes core principles of quantum computing, including qubits, superposition, entanglement, and measurement, through immersive interaction. Using a Mixed Reality headset, users engage with floating qubits, manipulate their states via controllers, and observe entanglement dynamics through real-time audiovisual feedback. A key feature of our implementation is the mathematically accurate and dynamic representation of qubits, both individually and while interacting with each other. The visualization of the qubit states evolve—rotate, shrink, grow, entangle—depending on their actual quantum states, which depend on variables such as proximity to other qubits and user interaction. Technical validation, along with preliminary demonstrations and feedback from quantum specialists, indicate that the system accurately represents quantum phenomena, suggesting strong potential to educate and enhance quantum intuition for non-expert audiences. A user study with non-experts further supports the system's accessibility and effectiveness in facilitating intuitive understanding of key quantum concepts. This approach bridges abstract quantum mechanics with embodied learning, offering an intuitive and accessible way for users to explore quantum phenomena. Future work will focus on expanding multi-user interactions and refining the fidelity of quantum state visualizations.

Abstract:
Recent advances in 3D Gaussian Splatting (3DGS) have enabled real-time, high-fidelity novel view synthesis. However, rendering each frame independently in a video sequence leads to redundant computations, especially when adjacent frames share significant visual overlaps. This inefficiency is particularly problematic in VR applications, where high frame rates and stereoscopic rendering amplify the per-frame cost. Existing frame interpolation or reuse strategies typically rely on image-domain information and are thus not directly applicable to 3DGS rendering, which is fundamentally point-based. To address this runtime inefficiency, we propose GSReuse, a lightweight and drop-in accelerator that speeds up 3DGS rendering by reusing computations across consecutive frames. GSReuse operates in screen space and introduces only minimal modifications to existing 3DGS rendering pipelines. It also eliminates the need for retraining scene representations. Given the rendered image, depth map, and camera parameters of the current frame, GSReuse estimates reliable Gaussian splatting motion vectors for all pixels and warps reusable contents to the new view. A tile-based filtering and masking strategy is then applied to determine which regions can be safely reused, allowing the 3DGS renderer to skip redundant rendering operations. We evaluate GSReuse on multiple benchmark datasets, showing that GSReuse significantly improves rendering, while maintaining high visual fidelity. Compared to state-of-the-art video frame reuse/generation methods, GSReuse delivers better image quality with much lower latency, facilitating practical deployment of 3DGS in VR applications.

Abstract:
Modeling the wireless radiance field (WRF) is fundamental to modern communication systems, enabling key tasks such as localization, sensing, and channel estimation. Traditional approaches, which rely on empirical formulas or physical simulations, often suffer from limited accuracy or require strong scene priors. Recent neural radiance field (NeRF)–based methods improve reconstruction fidelity through differentiable volumetric rendering, but their reliance on computationally expensive multilayer perceptron (MLP) queries hinders real-time deployment. To overcome these challenges, we introduce Gaussian splatting (GS) to the wireless domain, leveraging its efficiency in modeling optical radiance fields to enable compact and accurate WRF reconstruction. Specifically, we propose SwiftWRF, a deformable 2D Gaussian splatting framework that synthesizes WRF spectra at arbitrary positions under single-sided transceiver mobility. SwiftWRF employs CUDA-accelerated rasterization to render spectra at over 100 k FPS and uses the lightweight MLP to model the deformation of 2D Gaussians, effectively capturing mobility-induced WRF variations. In addition to novel spectrum synthesis, the efficacy of SwiftWRF is further underscored in its applications in angle-of-arrival (AoA) and received signal strength indicator (RSSI) prediction. Experiments conducted on both real-world and synthetic indoor scenes demonstrate that SwiftWRF can reconstruct WRF spectra up to 500x faster than existing state-of-the-art methods, while significantly enhancing its signal quality.

Abstract:
Biharmonic coordinates have become a powerful tool for cage-based deformation, owing to their inherent interpolation properties. However, their derivation for polynomial cages in 3D has remained unsolved. To address this, we propose closed-form expressions for polynomial 3D biharmonic coordinates and their derivatives when deformed from polygonal cages using the high-order boundary element method. Our primary contribution lies in the analytical derivation of the kernel integration using recursive differentiation techniques. Due to the enriched deformation space of biharmonic coordinates and the flexibility of polynomial cages, our method supports a broad range of deformations, as demonstrated through extensive experiments.

Abstract:
Predicting 3D Semantic Scene Graphs (3DSSG) is vital for understanding complex scenes by constructing structured representations. Current methods struggle with significant granularity discrepancies among instances, often relying on features at a single scale, which hampers their ability to perceive and interact with differently sized instances. To tackle this challenge, we introduce GranSSG, a novel approach that integrates volumetric granular awareness into 3DSSG prediction. Central to GranSSG is the Volumetric Pooling block, which aggregates features from multiple instance volumes, enhancing the representation of instance patterns across different granularities. Complementing this, the Granularity Transformer block dynamically directs attention to instance features across various network layers, ensuring precise perception of instances regardless of their granularity. Furthermore, the Cross-Granularity Correlation Transformer block mitigates performance degradation in instance pair relationship prediction by adaptively fusing hybrid features from different granularities, providing a comprehensive representation of instance pairs. Extensive evaluations on the challenging 3DSSG benchmark demonstrate that GranSSG significantly enhances prediction performance, setting a new state-of-the-art in 3DSSG prediction.

Abstract:
We present X2Video, the first diffusion model for rendering photorealistic videos guided by a sequence of intrinsic channels including albedo, normal, roughness, metallicity, and irradiance, while supporting intuitive multi-modal controls with reference images and text prompts for both global and local regions. The intrinsic guidance allows accurate manipulation of color, material, geometry, and lighting, while reference images and text prompts provide intuitive adjustments in the absence of intrinsic information. To enable these functionalities, we extend the intrinsic-guided image generation model XRGB to video generation by employing a novel and efficient Hybrid Self-Attention, which ensures temporal consistency across video frames and also enhances fidelity to reference images. We further develop a Masked Cross-Attention to disentangle global and local text prompts, applying them effectively onto respective local and global regions. For generating long videos, our novel Recursive Sampling method incorporates progressive frame sampling, combining keyframe prediction and frame interpolation to maintain long-range temporal consistency while preventing error accumulation. To support the training of X2Video, we assembled a video dataset named InteriorVideo, featuring 1,154 rooms from 295 interior scenes, complete with reliable ground-truth intrinsic channel sequences and smooth camera trajectories. Both qualitative and quantitative evaluations demonstrate that X2Video can produce long, temporally consistent, and photorealistic videos guided by intrinsic conditions. Additionally, X2Video effectively accommodates multi-modal controls with reference images, global and local text prompts, and simultaneously supports editing on color, material, geometry, and lighting through parametric tuning. Upon acceptance, we will publicly release our model and dataset.

Abstract:
We study the reconstruction of high-quality 3D head avatars. Our goal is to reduce the reliance on dense capture data for most existing approaches, which limits their practicality. Recent advances address this using single or few input images by either training a prior model or fine-tuning multi-view diffusion models to generate pseudo training points. While they fall short in producing multi-view-consistent, high-fidelity results aligned with the input data. This motivates us to explore a more practical and user-friendly input setting. Modern smartphones such as Apple's Face ID already guide users to slowly rotate their heads in front of a single camera, enabling the capture of facial data across varying viewpoints with minimal effort. This simple and intuitive scanning motion has become a widely accepted user habit and provides sufficient geometric information—highlighting a natural opportunity for 3D head avatar creation from monocular videos of neutral expression. Under this settinng, we introduce R^22 Avatar, a lightweight and user-friendly framework for generating expressive 3D head avatars, which adopts a Reconstruction-by-Restoration strategy avoiding large-scale model pretraining while achieving high-quality animatable avatars. Specifically, a geometry-guided warping module first synthesizes coarse expression variaHons from the neutral input. Then, a restoration module refines the warped results by recovering highfrequency facial details, including mouth interior, with the help of a data-driven 2D animation prior. These restored images serve as supervision targets to optimize the final avatar. Experiments demonstrate that our method produces realistic avatars with improved expression diversity and view consistency compared to baseline approaches.

Abstract:
Gaussian Splatting demonstrates impressive results in multi-view reconstruction based on Gaussian explicit representations. However, the current Gaussian primitives only have a single view-dependent color and an opacity to represent the appearance and geometry of the scene, resulting in a non-compact representation. In this paper, we introduce a new method called SVGS (Spatially Varying Gaussian Splatting) that utilizes spatially varying colors and opacity in a single Gaussian primitive to improve its representation ability. We have implemented bilinear interpolation, movable kernels, and tiny neural networks as spatially varying functions. SVGS employs 2D Gaussian surfels as primitives, which significantly enhances novel-view synthesis while maintaining high-quality geometric reconstruction. This approach is particularly effective in practical applications, as scenes combining complex textures with relatively simple geometry occur frequently in real-world environments. Quantitative and qualitative experimental results demonstrate that all three functions outperform the baseline, with the best movable kernels achieving superior novel view synthesis performance on multiple datasets, highlighting the strong potential of spatially varying functions.

Abstract:
Massive Open Online Courses (MOOCs) make high-quality instruction accessible. However, the lack of face-to-face interaction makes it difficult for instructors to obtain feedback on learners’ performance and provide more effective instructional guidance. Traditional analytical approaches, such as clickstream logs or quiz scores, capture only coarse-grained learning outcomes and offer limited insight into learners’ moment-to-moment cognitive states. In this study, we propose COIVis, an eye-tracking-based visual analytics system that supports concept-level exploration of learning processes in MOOC videos. COIVis first extracts course concepts from multimodal video content and aligns them with the temporal structure and screen space of the lecture, defining Concepts of Interest (COIs), which anchor abstract concepts to specific spatiotemporal regions. Learners’ gaze trajectories are transformed into COI sequences, and five interpretable learner-state features—Attention, Cognitive Load, Interest, Preference, and Synchronicity—are computed at the COI level based on eye tracking metrics. Building on these representations, COIVis provides a narrative, multi-view visualization enabling instructors to move from cohort-level overviews to individual learning paths, quickly locate problematic concepts, and compare diverse learning strategies. We evaluate COIVis through two case studies and in-depth user-feedback interviews. The results demonstrate that COIVis effectively provides instructors with valuable insights into the consistency and anomalies of learners’ learning patterns, thereby supporting timely and personalized interventions for learners and optimizing instructional design.

Abstract:
3D Gaussian Splatting (3DGS) has attracted considerable attention due to its remarkable rendering efficiency and superior visual quality. However, typical implementations struggle with ultra-high-resolution imagery due to excessive GPU memory demands, a critical limitation particularly in domains such as aerial mapping and remote sensing, where fine-grained details are crucial for accurate 3D reconstruction. In this paper, we introduce InfiniGS, an enhanced Gaussian Splatting framework that scales up training image resolution while maintaining fidelity and efficiency, enabling reconstruction from ultra-high-resolution images without encountering out-of-memory (OOM) issues. Our framework consists of two modules; firstly, we propose a Multi-Resolution Training strategy that guides the optimization from recovering global structures toward fine-grained details, while mitigating erosion artifacts during resolution transitions. Secondly, we introduce an Image Block Partitioning and Asynchronous Optimization scheme, which significantly reduces GPU memory overhead. Most importantly, we reveal the scaling rule for the densification threshold of Gaussian primitives and the hyperparameters of the Adam optimizer during resolution transitions and image partitioning. This ensures consistently high rendering quality regardless of the multi-resolution factor or partitioning strategy employed. Extensive experiments demonstrate that InfiniGS achieves superior Novel View Synthesis (NVS) for high-resolution scenes, while requiring less memory and shorter training time. To the best of our knowledge, InfiniGS is the first method capable of reconstructing scenes with resolutions up to 10 K and beyond, at full resolution and high fidelity, without exceeding the memory capacity of mainstream powerful GPUs (e.g., NVIDIA RTX 3090 with 24 GB VRAM).

Abstract:
Surface normal estimation is a fundamental task in point cloud processing and plays a crucial role in downstream applications. Existing methods typically extract features from local neighborhoods or patches, followed by surface fitting or direct regression to predict normals. However, the scale ambiguity in determining the optimal neighborhood hinders effective extraction of geometric information, making normal estimation for unstructured point clouds with significant density variations particularly challenging. To address this challenge, we propose APN-Net, an adaptive perception network for point cloud normal estimation. Specifically, we design the Graphical Information Self-perception (GIS) module, which provides an implicit manner for region partitioning and expands the receptive field, enabling automatic extraction of both local geometric details and global structural information, while alleviating the scale ambiguity in determining the optimal neighborhood. Moreover, to capture complex geometric details, we introduce the Adaptive Graph Convolution (AGC) module, which employs adaptive kernels to model relationships among points across different semantic regions, thereby enabling richer feature representation. Extensive experiments on both synthetic and real-world scanned datasets demonstrate that APN-Net achieves superior performance in unoriented normal estimation, particularly for point clouds with significant density variations.

Abstract:
We present a hierarchical triplet-based indoor relationship learning method, coined HierRelTriple, with a focus on spatial relationship learning. Existing approaches often depend on attention-based network designs and manually defined training objectives using handcrafted spatial rules or simplified pairwise relationships. However, these methods fail to capture complex, multi-object relationships found in real scenarios, leading to overcrowded or physically implausible arrangements. We introduce HierRelTriple, a hierarchical framework for modeling relational triplets, which first partitions functional regions and then automatically extracts three levels of spatial relationships: object-to-region (O2R), object-to-object (O2O), and corner-to-corner (C2C). By representing these relationships as geometric triplets and employing approaches based on Delaunay Triangulation to establish spatial priors, we derive IoU-based losses between denoised and ground-truth triplets and integrate them seamlessly into the diffusion denoising process. The joint formulation of inter-object distances, angular orientations, and spatial relationships enhances the physical realism of the generated scenes. Extensive experiments on unconditional layout synthesis, floorplan-conditioned layout generation, and scene rearrangement demonstrate that HierRelTriple improves spatial-relation metrics by over 15% and substantially reduces collisions and boundary violations compared to state-of-the-art methods.

Abstract:
The realistic and controllable generation of pure smoke is critical for smoke image editing, smoke visual special effects generation, and smoke data synthesizing within security scenarios. It is a relatively underexplored topic and continues to present significant challenges. Existing methods face challenges in the generation of smoke with intricate details and the regulation of various smoke styles. In this paper, a Pure Smoke image Generation Network (PSGNet) is proposed with a gradient and style learning approach to generate realistic and controllable smoke images. To achieve flexibility in control across the spatial dimension, the smoke shape mask is used to encode spatial details, such as the location and contour of the smoke, along with other related properties. To enhance the physical realism of synthesized smoke, a novel gradient-based learning framework is proposed to generate smoke gradient features, highlighting a special focus on explicitly encoding and exploiting gradient information. This framework uses a smoke gradient learning architecture that captures the subtle structures and patterns characteristic of real smoke, enabling the generation of highly realistic smoke with rich, fine-scale detail. In addition, a spatially aware style learning strategy is proposed to provide fine-grained control over smoke attributes such as density, color, and overall look. It is able to effectively model style features across both channel and spatial dimensions, thereby enabling spatially aware style manipulation. By combining the gradient module with this style learning framework, the method produces smoke that exhibits rich visual details and customizable image styles. Experiments conducted on six benchmark datasets demonstrate that the proposed PSGNet significantly outperforms the state-of-the-art approaches.

Abstract:
Weight perception is essential for delivering compelling and realistic object interaction in VR systems. While existing pseudo-haptic techniques have enabled users to perceive virtual object weight without physical actuation, their application has primarily been limited to near-field, direct hand interaction. As VR systems continue to advance in fidelity and versatility, interactions with objects beyond arm's reach are becoming increasingly common. However, how weight perception can be introduced into such remote interactions, and whether it can enhance user immersion and experience, remains underexplored. To bridge this gap, we present RayFlex, a pseudo-haptic technique that conveys object weight in raycasting-based interaction through visual displacement and dynamic ray deformation. The technique was evaluated in two user studies, which examined its effectiveness in supporting weight discrimination and its impact on user experience across different interaction contexts. Results indicate that RayFlex leads to significant improvements in perceived realism, presence, and satisfaction, while maintaining usability. From the results, we derived two implications and an application guideline that can help design future VR systems.

Abstract:
The visualization of temporal data on urban buildings, such as shadows, noise, and solar potential, plays a critical role in the analysis of dynamic urban phenomena. However, in dense and geographically constrained 3D urban environments, visual representations of time-varying building data often suffer from occlusion and visual clutter. To address these two challenges, we introduce an immersive lens visualization that integrates i) a view-dependent cutaway de-occlusion technique and ii) a temporal display derived from a conformal mapping algorithm. The mapping process first partitions irregular building footprints into smaller, sufficiently regular subregions that serve as structural primitives. These subregions are then seamlessly recombined to form a conformal, layered layout for our temporal lens visualization. The view-responsive cutaway is inspired by traditional architectural illustrations, preserving the overall layout of the building and its surroundings to maintain users' sense of spatial orientation. This lens design enables the occlusion-free embedding of shape-adaptive temporal displays across building facades on demand, supporting rapid time-space association for the discovery, access and interpretation of spatiotemporal urban patterns. Guided by domain and design goals, we outline the rationale behind the lens visual and interaction design choices, such as the encoding of time progression and temporal values in the conforming lens image. A controlled user study compares our approach against conventional juxtaposition and x-ray spatiotemporal designs. Results validate the usage and utility of our lens, showing that it improves task accuracy and completion time, reduces navigation effort, and increases user confidence. From these findings, we distill design recommendations and promising directions for future research on spatially-embedded lenses in 3D visualization, urban analytics, and related domains.

Abstract:
Virtual reality (VR) allows users to observe and manipulate 3D geometry from multiple viewpoints. Most VR selection work, however, optimizes techniques for selecting entire objects. Selecting a single face on a polyhedron remains underexplored and is more challenging because the interaction must act on a local component while preserving the object's global structure. We introduce a design space tailored to this task with three dimensions: viewing strategy, disambiguation consistency, and interaction metaphor. Guided by this space, we design eight freehand techniques for polyhedral face selection. A within-subjects study with 16 participants evaluates these techniques across polyhedral complexity (two radii; face counts 4, 6, and 12). The results identify three complementary top techniques, reveal tradeoffs between viewing choices and geometric preservation, and yield concrete guidelines for matching techniques to target geometry and task demands. A follow-up study with complex, realistic models confirms the robustness and practical usability of the three techniques. Together, these contributions shift attention from whole object selection to precise component selection in VR and provide actionable methods for 3D modeling, assembly, and texturing.

Abstract:
Music in virtual environments has long been treated as a temporal evolving element, enhancing atmosphere and game pace but rarely considered as a performance adaptive element. Recent advances in artificial intelligence (AI) and procedural audio make it possible to generate music that adapts in real time to player actions and system state. Yet, despite its potential, the behavioural impact of such adaptive generative soundtracks in head-mounted display-based virtual reality (VR) remains largely unexplored. To address this gap, we introduce a VR archery system that integrates Google MusicFXDJ with Ubiq-Genie to deliver continuous AI-generated adaptive music driven by gameplay events. In a within-subjects experiment (N = 22), participants completed trials with either a stylistically-matched fixed soundtrack or an adaptive soundtrack that escalated tension across four phases as arrows depleted. Measures combined self-reported ratings of presence, focus, stress, and emotional impact, with performance metrics of accuracy and aiming time. Results reveal that adaptive generative music not only heightens immersion and emotional salience but also modulates motor precision in an arousal-dependent inverted-U pattern: moderate musical tension improved accuracy and speed, whereas excessive tension impaired them. These findings establish AI-generated music as a powerful behavioural feedback modality in VR, opening pathways for training, rehabilitation, and next-generation immersive entertainment.

Abstract:
Virtual reality (VR) is increasingly influencing how researchers study human behavior. Yet, running large-scale VR experiments remains a challenge. The software is complex, capturing data from many sources at once is difficult, and analysis at scale takes significant effort. This paper reviews 67 open-source, research-oriented toolkits that support some or all stages of running VR studies, including the planning, design and management of experiments, collecting and synchronizing data, analyzing and visualizing results, and even running studies remotely or across different types of devices. We group existing peer-reviewed tools into five categories and highlight trends in the field. We discuss Unity's strong influence, the ongoing balance between easy-to-use interfaces and flexible coding approaches, the increased adoption of Lab Streaming Layer (LSL) for time-synchronized data capture, and growing interest in real-world and AI-supported analysis. Based on this overview, we offer practical recommendations to make VR research more consistent and sustainable, such as using shared formats for data and events, ensuring accurate timing and replay, and developing data exports that remain usable over time. We also compile a maintenance snapshot, which highlights technological and operational risks. We conclude with a roadmap targeting reproducibility, including considerations for a cross-platform VR experiment description language and distributed, replayable datasets. This review aims to help researchers choose and combine tools more effectively and give developers directions for supporting a more robust and reliable VR research ecosystem.

Abstract:
Cybersickness, a motion sickness like discomfort, is a major barrier to the usability of virtual reality (VR) systems. While prior work has focused mainly on predicting cybersickness severity, practical mitigation requires not only detecting how sick a user feels but also deciding whether a countermeasure is beneficial and determining its appropriate intensity. In this paper, we propose a two phase multitask learning framework that jointly models cybersickness severity, blur effectiveness, and blur intensity. In Phase 1, we pretrain temporal deep learning backbones on two single label datasets with only severity annotations. In Phase 2, we pro-gressively finetune the models on a multi-label dataset containing severity, blur effectiveness, and blur level labels. We evaluate three backbone architectures a Time-Series Transformer, Deep Temporal Convolutional Network, and TS-Mamba under a 10-fold block aware cross validation scheme. Results show that two phase training significantly outperforms single phase baselines, with the Time Series Transformer achieving best performance (FMS MAE = 0.57, R2 = 0.87; Blur Level MAE = 0.49, 2 = 0.95; Blur Preference ACC = 99.5%). Unlike prior rule based reduction frameworks that rely on static heuristics, our approach provides a data-driven “detect-decide-dose” pipeline that adapts blur mitigation dynamically to individual users. This demonstrates that single label pre-training is an effective strategy for developing multitask VR safety models under limited labeled data. To our knowledge, this is the first framework that unifies cybersickness prediction and adaptive reduction in a single model.

Abstract:
The aviation industry faces increasing pilot training demands, and reliance on conventional Full Flight Simulators (FFS) limits training capacity. Virtual Reality (VR) offers scalable, remote training opportunities, but its role as a complement to FFS requires empirical validation. In collaboration with Singapore Airlines (SIA) instructor pilots, we developed a VR training prototype for a visual approach into Gimhae International Airport, emphasizing Competency-Based Training and Assessment (CBTA)-based Threat and Error Management (TEM). An empirical study with 39 SIA Boeing 737-MAX 8 type-rated first officers evaluated VR against FFS. An equivalence analysis showed that VR achieved performance outcomes comparable to FFS in 13 of the 16 Observable Behaviors (OBs) and across 4 Competencies. In addition, a comparative analysis indicated measurable performance improvements when VR was used to supplement FFS training. Our results suggest VR can meaningfully complement FFS in targeted competency areas, with future work required to assess its broader integration across additional scenarios and performance metrics.

Abstract:
Despite the remarkable process of talking-head-based avatar-creating solutions, directly generating anchor-style videos with full-body motions remains challenging. In this study, we propose Make-Your-Anchor+, a novel system necessitating only a one-minute video clip of an individual for training, subsequently enabling the automatic generation of anchor-style videos with precise torso and hand movements. Specifically, we finetune a proposed structure-guided diffusion model on input video to render 3D mesh conditions into human appearances. We adopt a two-stage training strategy for the diffusion model, effectively mapping movements with specific appearances to create digital avatars for online streamers, live shopping hosts, and other applications. To produce arbitrary long temporal video, we extract human motion information from video diffusion prior by adapting the frame-wise diffusion model to pretrained video diffusion weights with lower cost, and a simple yet effective batch-overlapped temporal denoising module is proposed to bypass the constraints on video length during inference. Finally, a novel identity-specific face enhancement module is introduced to improve the visual quality of facial regions in the output videos. Comparative experiments demonstrate the system’s effectiveness and superiority in visual quality, temporal coherence, and identity preservation, outperforming SOTA diffusion/non-diffusion methods.

Abstract:
We explore the potential of visualization to support musicians in instrument practice through real-time feedback and reflection on their playing. Musicians often struggle to observe patterns in their playing and interpret them with respect to their goals. Our premise is that these patterns can be made visible with interactive visualization: we can make the unhearable visible. However, understanding the design of such visualizations is challenging: the diversity of needs, including different instruments, skills, musical attributes, and genres, means that any single use case is unlikely to illustrate the broad potential and opportunities. To address this challenge, we conducted a design exploration where we created and iterated on 33 designs, each focusing on a subset of needs, for example, only one musical skill. Our designs are grounded in our own experience as musicians and the ideas and feedback of 18 musicians with various musical backgrounds and we evaluated them with 13 music learners and teachers. This paper presents the results of our exploration, focusing on a few example designs as instances of possible instrument practice visualizations. From our work, we draw design considerations that contribute to future research and products for visual instrument education.

Abstract:
We present ForceCtrl, a novel 3D hand raycasting technique that enhances pointing precision based on control-display (CD) gain controlled with user-defined pinch force. We introduce a target-agnostic approach for refining raycasting precision, overcoming limitations in human motor accuracy. User-defined pinch force, detected with surface electromyography (sEMG), enables users to easily activate or deactivate CD gain during interaction. We propose three CD gain strategies and compare them through target selection and placement tasks. Our system reduces selection errors, placement jitters, and user workload, especially for distant targets in high-difficulty tasks. These results highlight the effectiveness of applying CD gain to hand raycasting and demonstrate the potential of user-defined pinch force as a robust input modality for precise hand interaction in AR/VR.

Abstract:
Singing-driven 3D head animation is a compelling yet underexplored task with broad applications in virtual avatars, entertainment, and education. Existing speech-driven approaches, which typically map audio directly to motion through implicit phoneme-to-viseme correspondences, often yield over-smoothed, emotionally flat, and semantically inconsistent results. These limitations render them inadequate for the unique demands of singing-driven animation. To address this challenge, we propose Think2Sing, a unified diffusion-based framework that integrates pretrained large language models to generate semantically consistent and temporally coherent 3D head animations conditioned on both lyrics and acoustics. Central to our framework is the introduction of motion subtitles, a structured, time-aligned representation generated via a Singing Chain-of-Thought process with acoustic-guided retrieval. These subtitles provide region-specific expressive cues that serve as interpretable priors for animation synthesis. We further formulate head animation as motion intensity prediction over key facial regions, enabling fine-grained control and more faithful expressive modeling. To support this paradigm, we construct the first multimodal singing dataset with synchronized 3D motion, acoustic descriptors, and aligned motion subtitles, enabling semantically grounded and expressive motion learning. Extensive experiments demonstrate that Think2Sing significantly outperforms state-of-the-art methods in realism, expressiveness, and emotional fidelity. Furthermore, our framework supports flexible subtitle-conditioned editing, enabling precise and user-controllable animation synthesis.

Abstract:
Radiance field techniques have demonstrated remarkable performance in reconstructing photorealistic real- world scenes. However, a widely recognized limitation of these methods is their reliance on densely captured images for training. Active learning offers a promising solution by selecting the most informative images to capture. An effective measure of data informativeness is the mutual information between the unknown image and the model parameters, known as the expected information gain. Naive estimation of this quantity requires inferring both the model and predictive distributions, which is computationally intractable for high-dimensional parameter spaces. In this work, we propose a computationally tractable method to estimate information gain. Instead of sampling in the high-dimensional model parameter space, we leverage the ray sampling process inherent in volume rendering to approximate the expected information gain. Specifically, we parameterize volume sampling with perturbed ray directions and learn the predictive distribution to infer optimal perturbation patterns. Furthermore, we derive an empirical risk decomposition that demonstrates how our method effectively explores the volume sampling space to enhance diversity, leading to an efficient and informative view selection. Experiments show that our approach achieves state-of-the-art performance in image quality for active learning of radiance fields, outperforming previous methods across various datasets, including both forward-facing and object-centric scenes.

Abstract:
Cell state discovery is crucial for understanding biological systems and enhancing medical outcomes. A key aspect of this process is identifying distinct biomarkers that define specific cell states. However, difficulties arise from the co-discovery process of cell states and biomarkers: biologists often use dimensionality reduction to visualize cells in a two-dimensional space. Then they usually interpret visually clustered cells as distinct states, from which they seek to identify unique biomarkers. However, this assumption is often this assumption often fails to hold due to internal inconsistencies in a cluster, making the process trial-and-error and highly uncertain. Therefore, biologists urgently need effective tools to help uncover the hidden association relationships between different cell populations and their potential biomarkers. To address this problem, we first designed a machine-learning algorithm based on the Mixture-of-Experts (MoE) technique to identify meaningful associations between cell populations and biomarkers. We further developed a visual analytics system–CellScout–in collaboration with biologists, to help them explore and refine these association relationships to advance cell state discovery. We validated our system through expert interviews, from which we further selected a representative case to demonstrate its effectiveness in discovering new cell states.

Abstract:
Virtual human animations have a wide range of applications in virtual and augmented reality. While automatic generation methods of animated virtual humans have been developed, assessing their quality remains challenging. Recently, approaches introducing task-oriented evaluation metrics have been proposed, leveraging neural network training. However, quality assessment measures for animated virtual humans not generated with parametric body models have yet to be developed. In this context, we introduce a first such quality assessment measure leveraging a novel data-driven framework. First, we generate a dataset of virtual human animations together with their corresponding subjective realism evaluation scores collected with a user study. Second, we use the resulting dataset to learn predicting perceptual evaluation scores. Results indicate that training a linear regressor on our dataset results in a correlation of 90%, which outperforms a strong deep learning baseline.

Abstract:
Nonlinear dimensionality reduction (NDR) techniques are widely used to visualize high-dimensional data. However, they often lack explainability, making it challenging for analysts to relate patterns in projections to original high-dimensional features. Existing interactive methods typically separate user interactions from the feature space, treating them primarily as post-hoc explanations rather than integrating them into the exploration process. This separation limits insight generation by restricting users’ understanding of how features dynamically influence projections. To address this limitation, we propose a bidirectional interaction method that directly bridges the feature space and the projections. By allowing users to adjust feature weights, our approach enables intuitive exploration of how different features shape the embedding. We also define visual semantics to quantify projection changes, enabling structured pattern discovery through automated query-based interaction. To ensure responsiveness despite the computational complexity of NDR, we employ a neural network to approximate the projection process, enhancing scalability while maintaining accuracy. We evaluated our approach through quantitative analysis, assessing accuracy and scalability. A user study with a comprehensive visual interface and case studies demonstrated its effectiveness in supporting hypothesis generation and exploratory tasks with real-world data. The results confirmed that our approach supports diverse analytical scenarios and enhances users’ ability to explore and interpret high-dimensional data through interactive exploration grounded in the feature space.

Abstract:
Creating high-quality line art in a fast and controlled manner plays a crucial role in anime production and concept design. We present DoodleAssist, an interactive and progressive line art generation system controlled by sketches and prompts, which helps both experts and novices concretize their design intentions or explore possibilities. Built upon a controllable diffusion model, our system performs progressive generation based on the last generated line art, synthesizing regions corresponding to drawn or modified strokes while keeping the remaining ones unchanged. To facilitate this process, we propose a latent distribution alignment mechanism to enhance the transition between the two regions and allow seamless blending, thereby alleviating issues of region incoherence and line discontinuity. Finally, we also build a user interface that allows the convenient creation of line art through interactive sketching and prompts. Qualitative and quantitative comparisons against existing approaches and an in-depth user study demonstrate the effectiveness and usability of our system. Our system can benefit various applications such as anime concept design, drawing assistant, and creativity support for children.

Affiliations: State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, Beijing, China; Computer School, Beijing Information Science and Technology University, Beijing, China; State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, and Zhongguancun Laboratory, Beijing, China; State Key Laboratory of Virtual Reality Technology and Systems, Beihang University, and Research Unit of Virtual Human and Virtual Surgery (RU), Chinese Academy of Medical Sciences, Beijing, China

Abstract:
Accurately reconstructing high-fidelity human models from single images is critical for virtual reality applications. Existing methods often rely on 3D features from the estimated parametric human model to provide geometric priors. This approach addresses challenges such as missing limbs or deformations, which often arise due to viewpoint limitations and self-occlusion. However, accurately predicting 3D features from monocular images remains a significant challenge. This limitation poses difficulties for the fusion of 2D and 3D information. In this paper, we introduce HFHuman, a novel approach for high-fidelity human reconstruction from a single image using multi-modality fusion. HFHuman effectively fuses multiple modalities, including geometric and depth, directly from images. Our method introduces three key innovations: (1) a depth and geometric parallel reconstruction framework that simultaneously handles whole-body geometry and detailed depth reconstruction, refining a parameterized 3D human model under progressive depth guidance; (2) a pixel-voxel feature fusion strategy that combines pixel-aligned features with voxel-aligned features using a multi-modality adaptor; and (3) a depth-refined technique that integrates RGB imagery with surface normals and depth mapping. By addressing the challenge of blending 2D and 3D modalities, HFHuman results in more accurate and realistic human reconstructions. Experimental results demonstrate that HFHuman outperforms state-of-the-art methods, setting a new standard for realistic 3D human body reconstruction.

Abstract:
With the growing demand for real-world 3-D understanding, learning effective representations of 3-D data has become increasingly important for tasks such as shape classification, model retrieval, scene reconstruction, and point cloud completion. Although previous work has explored self-supervised learning within individual modalities (e.g., point clouds or images), the potential of multi-modal supervision remains largely underexplored due to the lack of aligned and scalable training signals. In this work, we present DR-Point, a tri-modal pre-training framework that jointly learns from RGB images, depth maps, and 3-D point clouds to build a unified embedding space across modalities. By enforcing cross-modal consistency among RGB-depth-point triplets, DR-Point achieves effective 2-D-3-D feature alignment without manual annotations. A differentiable rendering module further enhances geometric fidelity by synthesizing depth cues and refining structural details in reconstructed point clouds. Extensive experiments on benchmarks demonstrate that DR-Point consistently outperforms state-of-the-art self-supervised methods on 3-D classification, segmentation, and completion. These results highlight the advantages of multi-modal pre-training for unified 3-D understanding and its potential to benefit a wide range of vision and graphics applications.

Abstract:
Scene synthesis is crucial to computer graphics. However, the current interactive scene synthesis methods usually cost the user too much time and interactions to edit objects. This paper presents a new interactive scene synthesis method that alleviates the designer from interacting with the 3D scene and its objects. Designers only need to select an object group in an independent panel through coarse clustering and fine clustering. Then, the furniture objects will be automatically added to the scene. This paper proposes a two-level clustering that applies the Affinity Propagation Algorithm (APA) to groups of furniture objects such that the object groups can even be clustered without linear representations, latent encoding, etc. To fully apply the APA, we also propose quantitatively measuring how different the two layouts are, i.e., how quantitatively the arrangements of two object groups differ. Experiments first show that our method is more user-friendly and interactively efficient than other interactive synthesis methods. By comparing our method with recent automatic scene synthesis methods, we demonstrate that our methods still have competitive plausibility. We also verify that our method does not harm the diversity and generalization of 3D scenes.

Abstract:
Understanding how dynamic patterns vary across large spatiotemporal ensembles is essential in many scientific domains. In fluid dynamics, for instance, researchers analyze how splash patterns in droplet impact experiments change with physical parameters such as fluid type or impact velocity. These experiments produce large volumes of data where patterns differ in size, shape, and duration, making manual analysis tedious and error-prone. Recently, interactive visualization approaches have been developed to assist analysis using learned similarity models for pattern-based querying. However, they assume fixed-size inputs and only support single-pattern queries, thus limiting their effectiveness for multiscale, multi-pattern analysis and exploration of ensembles. In this paper, we present a visual analysis approach for the interactive exploration of spatiotemporal ensembles through multiscale pattern querying. Our approach extends an existing similarity model to support variable-sized patterns, allowing users to define queries by selecting examples directly on visualized data. Coordinated views enable interactive querying, comparison, and analysis of pattern occurrences and relate pattern occurrences to ensemble parameters. A guidance mechanism supports the user in finding underexplored regions. We demonstrate the utility of our approach on synthetic and real-world datasets. Domain expert feedback confirms that the approach is intuitive, easy to use, and effective for revealing parameter-pattern relationships.

Abstract:
Novel view synthesis is a crucial task in the field of visualization and computer graphics. Although current methods, such as neural radiance fields (NeRF), have proven versatile for novel view synthesis, they often assume that input views are uniformly distributed in space. This means that each view captures a portion of the real-world scene, with each area being equally represented in the input views. In contrast, real-world scenarios involve randomly captured images with an uneven spatial distribution of input views, where some views of the scene are observed more frequently than others. This discrepancy poses significant challenges for current methods, underscoring the need to address the issue of nonuniformity in practical applications. To tackle this challenge, we propose a new dataset and an effective approach. Our extensive dataset comprises diverse scenes, capturing images in various real-world environments. Unlike previous novel view synthesis datasets, which have uniform or near-uniform input view distributions, our dataset effectively reflects the nonuniform distribution of images in real-world settings. Additionally, we introduce a simple yet effective method that significantly enhances the performance of current techniques under nonuniform conditions without modifying their pipelines. We anticipate that our data and method will serve as a valuable source of inspiration for future research in this area.

Abstract:
Online user studies of visualizations, visual encodings, and interaction techniques are ubiquitous in visualization research. Yet, designing, conducting, and analyzing studies effectively is still a major burden. Although various packages support such user studies, most solutions address only facets of the experiment life cycle, make reproducibility difficult, or do not cater to nuanced study designs or interactions. We introduce reVISit 2, a software framework that supports visualization researchers at all stages of designing and conducting browser-based user studies. ReVISit supports researchers in the design, debug & pilot, data collection, analysis, and dissemination experiment phases by providing both technical affordances (such as replay of participant interactions) and sociotechnical aids (such as a mindfully maintained community of support). It is a proven system that can be (and has been) used in publication-quality studies-which we demonstrate through a series of experimental replications. We reflect on the design of the system via interviews and an analysis of its technical dimensions. Through this work, we seek to elevate the ease with which studies are conducted, improve the reproducibility of studies within our community, and support the construction of advanced interactive studies.

Abstract:
Sentiment contagion occurs when attitudes toward one topic are influenced by attitudes toward others. Detecting and understanding this phenomenon is essential for analyzing topic evolution and informing social policies. Prior research has developed models to simulate the contagion process through hypothesis testing and has visualized user-topic correlations to aid comprehension. Nevertheless, the vast volume of topics and the complex interrelationships on social media present two key challenges: (1) efficient construction of large-scale sentiment contagion networks, and (2) in-depth explorations of these networks. To address these challenges, we introduce a causality-based framework that efficiently constructs and explains sentiment contagion. We further propose a map-like visualization technique that encodes time using a horizontal axis, enabling efficient visualization of causality-based sentiment flow while maintaining scalability through limitless spatial segmentation. Based on the visualization, we develop CausalMap, a system that supports analysts in tracing sentiment contagion pathways and assessing the influence of different demographic groups. Furthermore, we conduct comprehensive evaluations—including two use cases, a task-based user study, an expert interview, and an algorithm evaluation—to validate the usability and effectiveness of our approach.

Abstract:
Accurate and reliable visualization of spatiotemporal sensor data such as environmental parameters and meteorological conditions is crucial for informed decision-making. Traditional spatial interpolation methods, however, often fall short of producing reliable interpolation results due to the limited and irregular sensor coverage. This paper introduces a novel spatial interpolation pipeline that achieves reliable interpolation results and produces a novel heatmap representation with uncertainty information encoded. We leverage imputation reference data from Graph Neural Networks (GNNs) to enhance visualization reliability and temporal resolution. By integrating Principal Neighborhood Aggregation (PNA) and Geographical Positional Encoding (GPE), our model effectively learns the spatiotemporal dependencies. Furthermore, we propose an extrinsic, static visualization technique for interpolation-based heatmaps that effectively communicates the uncertainties arising from various sources in the interpolated map. Through a set of use cases, extensive evaluations on real-world datasets, and user studies, we demonstrate our model's superior performance for data imputation, the improvements to the interpolant with reference data, and the effectiveness of our visualization design in communicating uncertainties.

Abstract:
We present a new comprehensive theory for explaining, exploring, and using pattern as a visual variable in visualization. Although patterns have long been used for data encoding and continue to be valuable today, their conceptual foundations are precarious: the concepts and terminology used across the research literature and in practice are inconsistent, making it challenging to use patterns effectively and to conduct research to inform their use. To address this problem, we conduct a comprehensive cross-disciplinary literature review that clarifies ambiguities around the use of “pattern” and “texture”. As a result, we offer a new consistent treatment of pattern as a composite visual variable composed of structured groups of graphic primitives that can serve as marks for encoding data individually and collectively. This new and widely applicable formulation opens a sizable design space for the visual variable pattern, which we formalize as a new system comprising three sets of variables: the spatial arrangement of primitives, the appearance relationships among primitives, and the retinal visual variables that characterize individual primitives. We show how our pattern system relates to existing visualization theory and highlight opportunities for visualization design. We further explore patterns based on complex spatial arrangements, demonstrating explanatory power and connecting our conceptualization to broader theory on maps and cartography. An author version and additional materials are available on OSF: osf.io/z7ae2.

Abstract:
A growing body of work on visualization affordances highlights how specific design choices shape reader takeaways from information visualizations. However, mapping the relationship between design choices and reader conclusions often requires labor-intensive crowdsourced studies, generating large corpora of free-response text for analysis. To address this challenge, we explored alternative scalable research methodologies to assess chart affordances. We test four elicitation methods from human-subject studies: free response, visualization ranking, conclusion ranking, and salience rating, and compare their effectiveness in eliciting reader interpretations of line charts, dot plots, and heatmaps. Overall, we find that while no method fully replicates affordances observed in free-response conclusions, combinations of ranking and rating methods can serve as an effective proxy at a broad scale. The two ranking methodologies were influenced by participant bias towards certain chart types and the comparison of suggested conclusions. Rating conclusion salience could not capture the specific variations between chart types observed in the other methods. To supplement this work, we present a case study with GPT-40, exploring the use of large language models (LLMs) to elicit human-like chart interpretations. This aligns with recent academic interest in leveraging LLMs as proxies for human participants to improve data collection and analysis efficiency. GPT-40 performed best as a human proxy for the salience rating methodology but suffered from severe constraints in other areas. Overall, the discrepancies in affordances we found between various elicitation methodologies, including GPT-40, highlight the importance of intentionally selecting and combining methods and evaluating trade-offs.

Abstract:
Connected scatterplots visualize time-series data by connecting the points on a scatterplot based on temporal sequence. Viewers are prone to misinterpret the direction of time in these visualizations, possibly because they encode time with an unexpected rule - along the connected line (TIME IS A LINE) instead of from left to right (RIGHT IS LATER) as conventional in line charts. In this paper, we use the connected scatterplot to illustrate a perspective on visualization comprehension centered around expectations of encoding rules. People have initial expectations of encoding rules for visualizations that can stem from conventional practices or metaphors, and these expectations have been recognized as a potential factor influencing visualization comprehension. We present three preregistered experiments (n = 1429 in total) demonstrating two kinds of design interventions to strengthen the correct expectation for time and testing their effectiveness in reducing errors for understanding realistic connected scatterplots. We found that visual treatments that suppress the incorrect chart-type expectation and directional cues that emphasize the correct expectation both led viewers to expect TIME IS A LINE more. An explicit directional cue (arrows), ideally redundantly encoded with another cue (trace-line effect or animation), was most effective for reducing misinterpretations. Our findings not only provide practical guidelines for designing connected scatterplots but also contribute theoretical insights to inform the design of novel visualizations that challenge interpretability by defying expectations.

Abstract:
Although data visualization is powerful for revealing patterns and communicating insights, creating effective visualizations requires familiarity with authoring tools and often disrupts the analysis flow. While large language models show promise for automatically converting analysis intent into visualizations, existing methods function as black boxes without transparent reasoning processes, which prevents users from understanding design rationales and refining suboptimal outputs. To bridge this gap, we propose integrating Chain-of-Thought (CoT) reasoning into the Natural Language to Visualization (NL2VIS) pipeline. First, we design a comprehensive CoT reasoning process for NL2VIS and develop an automatic pipeline to equip existing datasets with structured reasoning steps. Second, we introduce nvBench-CoT, a specialized dataset capturing detailed step-by-step reasoning from ambiguous natural language descriptions to finalized visualizations, which enables state-of-the-art performance when used for model fine-tuning. Third, we develop DeepVIS, an interactive visual interface that tightly integrates with the CoT reasoning process, allowing users to inspect reasoning steps, identify errors, and make targeted adjustments to improve visualization outcomes. Quantitative benchmark evaluations, two use cases, and a user study collectively demonstrate that our CoT framework effectively enhances NL2VIS quality while providing insightful reasoning steps to users.

Abstract:
Various computational approaches predict chromatin structure, yielding concrete models that position genomic loci in physical space and help reveal genome organization and function. While prior visualization research has explored data and task abstractions for genomics, the design space for depicting these three-dimensional (3D) genome models—and associated genome-mapped data—remains unclear. In this paper, we investigate the visualization of genomic data with a spatial component. First, we systematically survey how 3D genome models are used and depicted in computational biology. We analyze over 300 papers with figures that visualize 3D genomic data and categorize the methods for visual representation. From this survey, we derive a design space for visualizing 3D genome data, identifying common patterns and key properties such as representation, visual channels, and composition. We position these findings within an existing genomics visualization taxonomy, refining and extending existing classifications. Second, we augment Gosling, a declarative visualization grammar for genomics, to support 3D genomic data. Our integration enables expressive authoring of visualizations that connect traditional genome-mapped information with 3D genome models, emphasizing their spatial characteristics. To demonstrate its utility, we employ our extended grammar to recreate interactive examples, showcasing its ability to represent complex visual designs. Comprehensive examples and an interactive editor are available at 3d.gosling-lang.org.

Abstract:
Accurate 3D aortic construction is crucial for clinical diagnosis, preoperative planning, and computational fluid dynamics (CFD) simulations, as it enables the estimation of critical hemodynamic parameters such as blood flow velocity, pressure distribution, and wall shear stress. Existing construction methods often rely on large annotated training datasets and extensive manual intervention. While the resulting meshes can serve for visualization purposes, they struggle to produce geometrically consistent, well-constructed surfaces suitable for downstream CFD analysis. To address these challenges, we introduce AortaDiff, a diffusion-based framework that generates smooth aortic surfaces directly from CT/MRI volumes. AortaDiff first employs a volume-guided conditional diffusion model (CDM) to iteratively generate aortic centerlines conditioned on volumetric medical images. Each centerline point is then automatically used as a prompt to extract the corresponding vessel contour, ensuring accurate boundary delineation. Finally, the extracted contours are fitted into a smooth 3D surface, yielding a continuous, CFD-compatible mesh representation. AortaDiff offers distinct advantages over existing methods, including an end-to-end workflow, minimal dependency on large labeled datasets, and the ability to generate CFD-compatible aorta meshes with high geometric fidelity. Experimental results demonstrate that AortaDiff performs effectively even with limited training data, successfully constructing both normal and pathologically altered aorta meshes, including cases with aneurysms or coarctation. This capability enables the generation of high-quality visualizations and positions AortaDiff as a practical solution for cardiovascular research.

Abstract:
Unstructured meshes present challenges in scientific data analysis due to irregular distribution and complex connectivity. Computing and storing connectivity information is a major bottleneck for visualization algorithms, affecting both time and memory performance. Recent task-parallel data structures address this by precomputing connectivity information at runtime while the analysis algorithm executes, effectively hiding computation costs and improving performance. However, existing approaches are CPU-bound, forcing the data structure and analysis algorithm to compete for the same computational resources, limiting potential speedups. To overcome this limitation, we introduce a novel task-parallel approach optimized for heterogeneous CPU-GPU systems. Specifically, we offload the computation of mesh connectivity information to GPU threads, enabling CPU threads to focus on executing the visualization algorithm. Following this paradigm, we propose GPU-Aided Localized data structurE (GALE), the first open-source CUDA-based data structure designed for heterogeneous task parallelism. Experiments on two 20-core CPUs and an NVIDIA V100 GPU show that GALE achieves up to 2.7× speedup over state-of-the-art localized data structures while maintaining memory efficiency.

Abstract:
The visualization and analysis of street and pedestrian networks are important to various domain experts, including urban planners, climate researchers, and health experts. This has led to the development of new techniques for street and pedestrian network visualization, expanding possibilities for effective data presentation and interpretation. Despite their increasing adoption, there is no established design framework to guide the creation of these visualizations while addressing the diverse requirements of various domains. When exploring a feature of interest, domain experts often need to transform, integrate, and visualize a combination of thematic data (e.g., demographic, socioeconomic, pollution) and physical data (e.g., zip codes, street networks), often spanning multiple spatial and temporal scales. This not only complicates the process of visual data exploration and system implementation for developers but also creates significant entry barriers for experts who lack a background in programming. With this in mind, in this paper, we reviewed 45 studies utilizing street-overlaid visualizations to understand how they are applied in practice. Through qualitative coding of these visualizations, we analyzed three key aspects of street and pedestrian network visualization usage: their analytical purposes, the visualization approaches employed, and the data sources used in their creation. Building on this design space, we introduce StreetWeave, a declarative grammar for designing custom visualizations of multivariate spatial network data across multiple resolutions. We demonstrate how StreetWeave can be used to create various street-overlaid visualizations, enabling effective exploration and analysis of spatial data. StreetWeave is available at urbantk.org/streetweave.

Abstract:
We present the Critical Design Strategy (CDS)-a structured method designed to facilitate the examination of visualisation designs through reflection and critical thought. The CDS helps designers think critically and make informed improvements using heuristic evaluation. When developing a visual tool or pioneering a novel visualisation approach, identifying areas for enhancement can be challenging. Critical thinking is particularly crucial for visualisation designers and tool developers, especially those new to the field, such as studying visualisation in higher education. The CDS consists of three stages across six perspectives: Stage 1 captures the essence of the idea by assigning an indicative title and selecting five adjectives (from twenty options) to form initial impressions of the design. Stage 2 involves an in-depth critique using 30 heuristic questions spanning six key perspectives-user, environment, interface, components, design, and visual marks. Stage 3 focuses on synthesising insights, reflecting on design decisions, and determining the next steps forward. We introduce the CDS and explore its use across three visualisation modules in both undergraduate and postgraduate courses. Our longstanding experience with the CDS has allowed us to refine and develop it over time: from its initial creation through workshops in 2017/18 to improvements in wording and the development of two applications by 2020, followed by the expansion of support notes and refinement of heuristics through 2023; while using it in our teaching each year. This sustained use allows us to reflect on its practical application and offer guidance on how others can incorporate it into their own work.

Abstract:
Complex, high-dimensional data is ubiquitous across many scientific disciplines, including machine learning, biology, and the social sciences. One of the primary methods of visualizing these datasets is with two-dimensional scatter plots that visually capture some properties of the data. Because visually determining the accuracy of these plots is challenging, researchers often use quality metrics to measure the projection's accuracy and faithfulness to the original data. One of the most commonly employed metrics, normalized stress, is sensitive to uniform scaling (stretching, shrinking) of the projection, despite this act not meaningfully changing anything about the projection. Another quality metric, the Kullback–Leibler (KL) divergence used in the popular t-Distributed Stochastic Neighbor Embedding (t-SNE) technique, is also susceptible to this scale sensitivity. We investigate the effect of scaling on stress and KL divergence analytically and empirically by showing just how much the values change and how this affects dimension reduction technique evaluations. We introduce a simple technique to make both metrics scale-invariant and show that it accurately captures expected behavior on a small benchmark.

Abstract:
Indoor scene synthesis is essential for creative industries. Driven by this demand, recent advances in scene synthesis using diffusion and autoregressive models have shown promising results. However, existing models struggle to achieve real-time performance, high visual fidelity, and flexible scene editing simultaneously. To tackle this challenge, we propose MaskScene, a novel hierarchical conditional masked model for real-time 3D indoor scene synthesis and editing. Specifically, MaskScene introduces a hierarchical scene representation that explicitly encodes scene relationships, semantics, and tokens. Based on this representation, we design a hierarchical conditional masked modeling architecture that enables parallel and iterative decoding, conditioned on both semantics and relationships. MaskScene leverages local object masking and hierarchical scene structures to infer occluded regions from partial observations, enabling rapid construction of 3D indoor environments that accurately reflect real-world scenes. Compared to state-of-the-art methods, MaskScene achieves 80× faster generation speed and improves scene quality by 10%, while also supporting zero-shot editing, such as scene completion and rearrangement, without extra fine-tuning.

Abstract:
The generation of anchor-style product promotion videos presents promising opportunities in e-commerce, advertising, and consumer engagement. Despite advancements in pose-guided human video generation, creating product promotion videos remains challenging. In addressing this challenge, we identify the integration of human-object interactions (HOI) into pose-guided human video generation as a core issue. To this end, we introduce AnchorCrafter, a novel diffusion-based system designed to generate 2D videos featuring a target human and a customized object, achieving high visual fidelity and controllable interactions. Specifically, we propose two key innovations: the HOI-appearance perception, which enhances object appearance recognition from arbitrary multi-view perspectives and disentangles object and human appearance, and the HOI-motion injection, which enables complex human-object interactions by overcoming challenges in object trajectory conditioning and inter-occlusion management. Extensive experiments show that our system improves object appearance preservation by 7.5%, and achieves the best video quality compared to existing state-of-the-art approaches. It also outperforms existing approaches in maintaining human motion consistency and high-quality video generation.

Abstract:
Estimating the 6-DoF posture of parts in assembly-based modeling is a critical task in the fields of computer graphics, computer vision and robotics. A typical scenario involves enabling a machine agent to automatically assemble IKEA furniture using the provided parts. This paper presents HiFormer, a novel Hierarchical Transformer with Box-packed Positional Encoding, designed for highly automatic 3D part assembly. Our method addresses three important issues commonly encountered in 3D part assembly: 1) How to mitigate the overfitting problem associated with Transformer-based feature learning for 3D point clouds? 2) How to effectively model the relationships between the intragroup and intergroup parts? 3) How to compute positional encoding and integrate it into the Transformer for parts with diverse geometric forms in the coarse-to-fine assembly process? These challenges are tackled through three key contributions: 1) a multi-task 3D Swin Transformer with a two-stage training strategy for feature extraction, 2) a novel hierarchical Transformer for capturing part relationships at flattening, intragroup, and intergroup levels, and 3) an innovative box-packed positional encoding that enhances the Transformer by incorporating query, key, and value information derived from relative box positions. On the PartNet benchmark, our method outperforms the state-of-the-art PWH-MP model on three representative categories—Chair, Table, and Lamp—, achieving average improvements of 2.84% in Part Accuracy (PA) and 3.72% in Connection Accuracy (CA) for diversity modeling (with noise), and 3.55% in PA and 3.21% in CA for deterministic modeling (without noise).

Abstract:
Automated floor plan generation aims to create residential layouts by arranging rooms within a given boundary, balancing topological, geometric, and aesthetic considerations. The existing methods typically use a multi-step pipeline with intermediate representations to decompose the prediction process into several sub-tasks, limiting model flexibility and imposing predefined solution paths. This often results in unreasonable outputs when applied to data unsuitable for these predefined paths, making it challenging for these methods to match human designers, who do not restrict themselves to a specific set of design workflows. To address these limitations, we introduce CE2EPlan, a controllable end-to-end topology- and geometry-enhanced diffusion model that removes restrictions on the generative process of AI design tools. Instead, it enables the model to learn how to design floor plans directly from data, capturing a wide range of solution paths from input boundaries to complete layouts. Extensive experiments demonstrate that our method surpasses all existing approaches using the multi-step pipeline, delivering higher-quality results with enhanced user control and greater diversity in output, bringing AI design tools closer to the versatility of human designers.

Abstract:
This work introduces a new approach to automatic oil painting that emphasizes the creation of dynamic and expressive brushstrokes. A pivotal challenge lies in mitigating the duplicate and common-place strokes, which often lead to less aesthetic outcomes. Inspired by the human painting process, i.e., observing, comparing, and drawing, we incorporate differential image analysis into a neural oil painting model, allowing the model to effectively concentrate on the incremental impact of successive brushstrokes. To operationalize this concept, we propose the Differential Query Transformer (DQ-Transformer), a new architecture that leverages differentially derived image representations enriched with positional encoding to guide the stroke prediction process. This integration enables the model to maintain heightened sensitivity to local details, resulting in more refined and nuanced stroke generation. Furthermore, we incorporate adversarial training into our framework, enhancing the accuracy of stroke prediction and thereby improving the overall realism and fidelity of the synthesized paintings. Extensive qualitative evaluations, complemented by a controlled user study, validate that our DQ-Transformer surpasses existing methods in both visual realism and artistic authenticity, typically achieving these results with fewer strokes.

Abstract:
Does seeing avatars during social interaction in virtual reality (VR) affect the neural synchrony of interacting people? To address this question, we assessed inter-brain synchronization (IBS), which reflects synchronization of brain activities among interacting people and is linked to interaction success. We designed a social interaction task in VR where participants’ appearances were represented by virtual cursors only or additional human avatars on them. Participants performed a finger position tracking interaction under both conditions. We compared the IBS across interaction conditions by collecting electroencephalography (EEG), positional data, and virtual embodiment questionnaires. The results from 30 participants (15 pairs) show that seeing the avatars resulted in significantly higher virtual body ownership but comparable agency and positional synchronization compared to the virtual cursors-only condition. In the inter-brain analysis, we found a significant increase in IBS strength and an increasing trend in the number of IBS in the alpha band after interacting with avatar-embodied virtual cursors. Correlation analyses further revealed a significant association between alpha-band IBS and subjective body ownership, and a trend with social inclusion scores. While the influence of visual complexity cannot be ruled out, these findings suggest that avatar-induced embodiment may contribute to enhanced inter-brain synchrony during VR-mediated social interaction.

Abstract:
Reconstructing transparent objects with high fidelity presents significant challenges due to complex light refraction and reflection. Existing methods rely on intentionally designed patterns observed behind the transparent object to infer the correspondence between rays and the background, thereby improving the precision of the reconstruction. However, they are hindered by a refraction-tracing-based strategy that fails to reconstruct complex nested transparent objects and a tedious view-capture strategy relying on images captured from empirically determined viewpoints. To overcome these obstacles, we propose AHC-NeRF, an autonomous, high-quality neural SDF-based framework designed for reconstructing two-layer complex nested transparent objects. Firstly, our framework combines neural SDF with single-pixel imaging, a reflection-based method, which utilizes point-pair priors as guidance to achieve high-quality reconstruction of both the outer and inner surfaces. Secondly, we propose an adaptive single-pixel imaging method that achieves an acceleration of 1–2 orders of magnitude compared to vanilla single-pixel imaging for the acquisition of point-pair priors. Finally, we introduce a novel view-planning strategy that progressively identifies the viewpoints with the highest information gain throughout the optimization process, thereby achieving high-quality surface reconstruction. Extensive experimental results on both synthetic and real-world datasets demonstrate that AHC-NeRF outperforms state-of-the-art methods.

Abstract:
Surface reconstruction from point clouds has attracted considerable attention in the computer vision community. Most deep learning-based methods focus on modeling signed distance functions (SDFs) and unsigned distance functions (UDFs). However, SDFs are limited to watertight models, and UDFs are non-differentiable at the surface boundary, hindering the learning of smooth representations. In this paper, we introduce a novel method, VFIR, which leverages neural implicit functions to learn vector fields (VFs) for 3D shape modeling. We enhance the network’s fitting capability by displacing points along the predicted vector directions, aiming to reduce the deviation between the generated point cloud and the original input. To accelerate convergence, we introduce a progressive learning strategy that incrementally incorporates denser point clouds and corresponding normals during training. For isosurface extraction, we define truncated vector fields (TVFs) that focus computation around surfaces and present an optimized three-state marching cubes (OT-MC) algorithm tailored to the vector field representation. Extensive experiments demonstrate that VFIR achieves state-of-the-art performance with high accuracy and robustness across diverse 3D models, exhibiting strong generalization capabilities and thus representing a promising solution for surface reconstruction across various applications.

Abstract:
Vection, the visual illusion of self-motion, provides a strong marker of the VR user experience and plays an important role in both presence and cybersickness. Traditional measurements have been conducted using questionnaires, which exhibit inherent limitations due to their subjective nature and prevent real-time adjustments. Detecting vection in real time would allow VR systems to adapt to users’ needs, improving comfort and minimizing negative effects like cybersickness. This paper investigates the presence of vection markers in electroencephalographic (EEG) brain signals using evoked potentials (brain responses to external stimuli). We designed a VR experiment that induces vection using two conditions: (1) forward acceleration or (2) backward acceleration. We recorded electroencephalographic (EEG) signals and gathered subjective reports on thirty (30) participants. We found an evoked potential of vection characterized by a positive peak around 600 ms (P600) after stimulus onset in the parietal region and a simultaneous negative peak in the frontal region. This result paves the way for the automatic detection of vection using EEG as well as a better understanding of vection. It also provides insights into the functional role of the visual system and its integration with the vestibular system during motion-perception. It has the potential to help enhance VR user experience by qualifying users’ perceived vection and adapting the VR environments accordingly.

Abstract:
Dragging is a fundamental interaction technique in human-computer interaction. Prior research on dragging has generally been concerned with 2D objects on 2D interactive surfaces, with limited attention to 3D interactive spaces. We performed three experiments to systematically investigate dragging using raycasting in virtual reality (VR), a representative 3D interactive environment. The first experiment examined dragging 2D targets on 2D surfaces placed in VR. The second experiment investigated dragging 3D targets positioned at identical depths in VR. Both experiments examined the effects of target depth, angular target amplitude, and angular target on dragging performance. The third experiment investigated dragging 3D targets with varying depths, focusing on the effects of target layout, target amplitude, and target width. Findings revealed the different impacts of the above factors on dragging interaction. We also proposed a new 3D Fitts’ law model tailored for dragging using raycasting in VR, which fits well with our data. Finally, we discuss our results and provide design recommendations for dragging (i.e., translational positioning) tasks using raycasting under the tested experimental conditions in VR.

Abstract:
High-fidelity 3D Gaussian Splatting methods excel at capturing fine textures but often overlook model compactness, resulting in massive splat counts, bloated memory, long training, and complex post-processing. We present Micro-Splatting: Multistage Isotropy-informed Covariance Regularization Optimization for High-Fidelity 3D Gaussian Splatting, a unified pipeline that preserves visual detail while drastically reducing model complexity without any post-processing or auxiliary neural modules. In Stage I (Growth), we introduce a trace-based covariance regularization to maintain near-isotropic Gaussians, mitigating low-pass filtering in high-frequency regions and improving spherical-harmonic color fitting. We then apply gradient-guided adaptive densification that subdivides splats only in visually complex regions, leaving smooth areas sparse. In Stage II (Refinement), we prune low-impact splats using a simple opacity–scale importance score and merge redundant neighbors via lightweight spatial and feature thresholds, producing a lean yet detail-rich model. On three benchmarks, Micro-Splatting reduces splat count and model size by up to 60% and shortens training by 20%, while matching or surpassing state-of-the-art PSNR, SSIM, and LPIPS in real-time rendering. These results demonstrate that Micro-Splatting delivers both compactness and high fidelity in a single, efficient, end-to-end framework.

Abstract:
Virtual Reality (VR) co-manipulation enables multiple users to collaboratively interact with shared virtual objects. However, existing research treats objects as monolithic entities, overlooking scenarios where users need to manipulate different sub-components simultaneously. This work addresses conflict resolution when users select overlapping vertices (non-disjoint sets) during co-manipulation. We present a comprehensive framework comprising preventive strategies (Object-level and Action-level Restrictions) and reactive strategies (computational conflict resolution). Through two user studies with 76 participants (38 pairs), we evaluated these approaches in collaborative wireframe editing tasks. Study 1 identified Averaging as the optimal computational method, balancing task efficiency with user experience. Study 2 highlighted that Action-level Restriction, which permits overlapping selections but restricts concurrent identical operations, achieved better performance compared to exclusive object locking. Reactive strategies using averaging provided smooth collaboration for experienced users, while second-user priority enabled quick corrections. Our findings indicate that optimal strategy selection depends on task requirements, user expertise, and collaboration patterns. Based on the findings, we provide design implications for developing VR collaboration systems that support flexible sub-components manipulation while maintaining collaborative awareness and minimizing conflicts.

Abstract:
Merge trees are fundamental topological descriptors with broad theoretical and practical significance, yet their computation, particularly for augmented merge trees, incurs substantial computational and memory overhead. We present BiGMT, a highly scalable parallel algorithm for merge tree construction based on regions of the Morse-Smale segmentation. By exploiting boundary vertex characteristics and the dynamic merging of these regions, BiGMT preserves the essential topology of d-dimensional manifolds and represents the structure as a weighted bipartite graph. This formulation decomposes merge tree construction into fine-grained parallel tasks, significantly reducing traversal cost and memory consumption. Furthermore, BiGMT efficiently constructs augmented merge trees using reverse binary lifting search and parallel connection mechanisms, substantially improving augmentation performance. Extensive experiments on medium- to large-scale scalar field datasets demonstrate substantial GPU acceleration. BiGMT consistently outperforms ExTreeM, PPP, and FTM-Tree on RTX4090 and H200, achieving speedups of up to 23.95× and demonstrating superior parallel efficiency and scalability.

Abstract:
Audio-driven human gesture synthesis is a crucial task with broad applications in virtual avatars, human-computer interaction, and creative content generation. Despite notable progress, existing methods often produce coarse gestures, lack expressiveness, and fail to fully align with audio semantics. To address these challenges, we propose ExGes, a novel retrieval-enhanced diffusion framework with three key designs: (1) a Motion Base Construction, which builds a gesture library from the training dataset; (2) a Motion Retrieval Module, employing contrastive learning and momentum distillation for retrieving fine-grained reference poses; and (3) a Precise Control Module, integrating partial masking and stochastic masking to enable flexible and fine-grained control. Experimental evaluations on BEAT2 demonstrate that ExGes reduces Fréchet Gesture Distance by 4.55%and improves motion diversity by 5.3% over EMAGE, with user studies revealing a 71.3% preference for its naturalness and semantic relevance.

Abstract:
With increasing computational power, scientists often employ high-fidelity simulations to study complex scientific phenomena. However, these simulations remain slow and costly in terms of computation and storage, prompting the use of faster low-fidelity alternatives. Yet, the distribution gap between low- and high-fidelity data, due to missing fine-scale details and simplified physics, hampers accurate scientific understanding. To overcome these challenges, we propose a neural operator-based framework for multi-fidelity prediction on unstructured data. Our framework leverages a graph neural operator to effectively map low-fidelity data to the high-fidelity counterparts. It further incorporates a spectral-based module that captures fine-scale details, enhancing the reconstruction of high-fidelity fields. Extensive experiments across diverse datasets demonstrate that our method consistently surpasses strong baselines, including state-of-the-art neural operators and learning methods for unstructured data, highlighting its robustness and effectiveness in bridging fidelity gaps.

Abstract:
Finding a particular object in a display is important for viewers in many visualizations, for example, when reacting to brushing or to a highlighted object. This can be enabled by making the target object different in one of the visual variables that determine the object's appearance; for example, by changing its color or size. Certain interpretations of the visual search literature have promoted the view that using visual variables such as hue-often labeled as preattentive-would make the target object automatically “popout,” implying that an object can be located almost instantly, regardless of the number of objects in the display. In this paper we present a study that serves as a bridge between the extensive visual search literature and visualization, establishing empirical base measurements for the localization task. By testing displays with up to hundreds of objects, we are able to show that none of the common visual variables is immune to the increase in the number of objects. We also provide the first empirically informed comparisons between visual variables for this task in the context of visualization, and show how different visual variables have varying robustness with respect to two additional dimensions: the location of the target and the overall visual arrangement (layout). A free copy of this paper and all supplemental materials are available on our online repository: https://osf.io/z68ak/overview.

Abstract:
Coding agents powered by large language models (LLMs) have gained traction for automating code generation through iterative problem-solving with minimal human involvement. Despite the emergence of various frameworks, e.g., LangChain, AutoML, and AIDE, ML scientists still struggle to effectively review and adjust the agents' coding process. The current approach of manually inspecting individual outputs is inefficient, making it difficult to track code evolution, compare coding iterations, and identify improvement opportunities. To address this challenge, we introduce a visual analytics system designed to enhance the examination of coding agent behaviors. Focusing on the AIDE framework, our system supports comparative analysis across three levels: (1) Code-Level Analysis, which reveals how the agent debugs and refines its code over iterations; (2) Process-Level Analysis, which contrasts different solution-seeking processes explored by the agent; and (3) LLM-Level Analysis, which highlights variations in coding behavior across different LLMs. By integrating these perspectives, our system enables ML scientists to gain a structured understanding of agent behaviors, facilitating more effective debugging and prompt engineering. Through case studies using coding agents to tackle popular Kaggle competitions, we demonstrate how our system provides valuable insights into the iterative coding process.

Abstract:
Multi-camera dynamic Augmented Reality (AR) applications require a camera pose estimation to leverage individual information from each camera in one common system. This can be achieved by combining contextual information, such as markers or objects, across multiple views. While commonly cameras are calibrated in an initial step or updated through the constant use of markers, another option is to leverage information already present in the scene, like known objects. Another downside of marker-based tracking is that markers have to be tracked inside the field-of-view (FoV) of the cameras. To overcome these limitations, we propose a constant dynamic camera pose estimation leveraging spatiotemporal FoV overlaps of known objects on the fly. To achieve that, we enhance the state-of-the-art object pose estimator to update our spatiotemporal scene graph, enabling a relation even among non-overlapping FoV cameras. To evaluate our approach, we introduce a multi-camera, multi-object pose estimation dataset with temporal FoV overlap, including static and dynamic cameras. Furthermore, in FoV overlapping scenarios, we outperform the state-of-the-art on the widely used YCB-V and T-LESS dataset in camera pose accuracy. Our performance on both previous and our proposed datasets validates the effectiveness of our marker-less approach for AR applications. The code and dataset are available on https://github.com/roth-hex-lab/IEEE-VR-2026-MultiCam.

Abstract:
Immersive virtual reality (VR) applications demand accurate, temporally coherent full-body pose tracking. Recent head-mounted camera-based approaches show promise in egocentric pose estimation, but encounter challenges when applied to VR head-mounted displays (HMDs), including temporal instability, inaccurate lower-body estimation, and the lack of real-time inference. To address these limitations, we present EgoPoseVR, an end-to-end framework for accurate egocentric full-body pose estimation in VR that integrates headset motion cues with egocentric RGB-D observations through a dual-modality fusion pipeline. A spatiotemporal encoder extracts frame- and joint-level representations, which are fused via cross-attention to fully exploit complementary motion cues across modalities. A kinematic optimization module then imposes constraints from HMD signals, enhancing the accuracy and stability of pose estimation. To facilitate training and evaluation, we introduce a large-scale synthetic dataset of over 1.8 million temporally aligned HMD and RGB-D frames across diverse VR scenarios. Experimental results show that EgoPoseVR outperforms state-of-the-art egocentric pose estimation models. A user study in real-world scenes further shows that EgoPoseVR achieved significantly higher subjective ratings in accuracy, stability, embodiment, and intention for future use compared to baseline methods. These results show that EgoPoseVR enables robust full-body pose tracking, offering a practical solution for accurate VR embodiment without requiring additional body-worn sensors or room-scale tracking systems.

Abstract:
A convincing sense of embodiment in virtual reality (VR) is crucial for creating immersive and engaging experiences, as it shapes how users perceive and interact with their virtual bodies. The sense of embodiment is thereby, among others, affected by the shape, appearance, and fidelity of the virtual body. However, achieving convincing avatar appearance remains a challenge for VR applications. One promising solution is Pass-Through Embodiment (PTE), which enables users to see their real bodies in VR. PTE combines depth-based segmentation with the pass-through video stream of video-see-through displays to effectively visualize photon-captured representations of their own bodies. Despite the source-fidelity of the representation, the resolution of integrated depth sensors in Head Mounted Displays (HMD) can produce artifacts at segmentation boundaries, leading to visible aliasing. The perceptual impact of these artifacts on the VR experience remains unexplored. Therefore, in this paper we compare three edge-rendering techniques designed to reduce artifacts without compromising performance. Aside from a soft gradient, we introduce two new methods with a hard and dithered edge. The latter aims to balance the sharpness of hard masks with the smoothness of gradient transitions, without relying on alpha blending. To evaluate those methods in a PTE context, we conducted a within-subjects study that introduces a novel real-mirror paradigm, using an actual physical mirror as reference for reflection. We found significant results in measured presence and embodiment. Subsequent analysis revealed that our dithered cutout approach significantly outperforms hard masks, while no significant difference was found between soft condition. These results suggest a perceptual continuum where dithering and soft blending both effectively reduce visual artifacts through gradient representation. Together with high overall ratings on presence and embodiment across all conditions, these findings confirm PTE as a robust method for supporting embodiment and presence, while highlighting the potential of dithering as a computationally efficient yet perceptually comparable alternative to smooth blending.

Abstract:
Rule-breaking behaviors, such as jaywalking or skipping queues, are common in crowds but difficult to study in real-world settings due to limited control and observability. Virtual reality (VR) provides a controlled alternative, but its validity depends on whether VR elicits realistic rule-breaking behavior. We conducted a VR study with 65 participants navigating a virtual city with four scenarios differing in social norms: walking on grass, crossing outside a crosswalk, jaywalking at a red light, and skipping a line. In each scenario, the proportion of rule-breaking virtual characters was manipulated (0%, 25%, 50%). Participants' movements and gaze were recorded to assess behavior and attention. Results showed higher rule-breaking as the number of violators increased, except in the low-stakes crossing scenario. Rule-breakers attended more to violating characters, whereas rule-followers focused on compliant ones. Participants cited efficiency, safety, and social norms as key factors guiding their decisions. Overall, VR reproduced natural patterns of social compliance and noncompliance, supporting its use for studying rule-breaking and applications in crowd simulation, urban design, safety training, and immersive media.

Abstract:
Mixed Reality (MR) telecollaboration aims to enable users to share local objects as real-time synchronized virtual replicas to remote partners and collaborate on physical tasks as if they were co-located. However, in everyday scenarios with mobile and easy-to-setup MR devices, visualizing shared objects in a single modality, ranging from 2D images to 3D reconstruction, struggles to simultaneously optimize all the aspects of Spatiality, Fidelity, and Real-time performance. To overcome this issue, existing methods explore integrating multiple visualization modalities to leverage their respective advantages in subsets of the three aspects. However, they focus on fixed modality combinations without considering user-centered task contexts and workflow, where users may prioritize different aspects of the visualization across task phases. Moreover, they lack support for switching or require manual switching across modalities, which could become disruptive and tiring. In this paper, we propose adapting object visualization based on spatiotemporal contexts in telecollaboration. Specifically, we first couple task type with the user's relative viewing distance as the spatial context, and examine its impact on users' prioritized visualization aspects, and the corresponding switching thresholds. With differing generation speeds of modalities, we then explore temporal switching schemes when the preferred modality is not immediately available. With the obtained design choices, we implement CollabVisAdapt, a proof-of-concept prototype that supports automatic adaptation of object visualization based on spatiotemporal contexts in MR telecollaboration. A user study in remote maintenance verifies the effectiveness of the proposed workflow with adaptive visualization and the usability of the system.

Abstract:
Various locomotion techniques, such as teleportation, walking in place, redirected walking, and walking, have been proposed. However, conventional methods overlooked the human ability to coordinate multiple modalities (e.g., eyes and feet) for natural virtual locomotion within a small physical space. Inspired by the natural coordination of the eyes and feet in human walking, as well as insights from prior work, this study investigates how eye-foot coordination can be leveraged more effectively for VR teleportation. We present four novel teleportation techniques based on eye-foot coordination, using users' gaze behavior and 3D foot positions as input modalities. A user study with 20 participants compared our techniques with a state-of-the-art foot-based locomotion method. Results demonstrate the superiority of our approaches: task completion times were significantly reduced, NASA-TLX workload scores and SUS usability scores were markedly improved, and participants expressed a clear preference for our techniques over the baseline. These findings provide a strong foundation for the design and implementation of future eye-foot coordinated teleportation methods in VR.

Abstract:
The inherent freedom of exploration in virtual reality poses a challenge for directing user attention toward relevant points or objects of interest, which are potentially located outside the user's field of view or are temporarily obscured. Accordingly, guidance must sustain users' perceptual focus while preserving the sense of presence. This paper presents RippleVision, a subtle gaze guidance technique using wave-like ripples that modulate brightness with inverted polarity between the eyes and appear only within a cone from the point of interest to the gaze location. A calibration study defined detectability and acceptability thresholds for the cues, followed by a comparative user study against four state-of-the-art techniques. Results show RippleVision achieves similar search times to clearly visible techniques while significantly reducing visual dominance. Moreover, its effectiveness scales with cue visibility, improving search performance with minimal impact on perceived visual obstruction and presence.

Abstract:
Opportunities for medical students to gain practical experience in vaginal births are increasingly constrained by shortened clinical rotations, patient reluctance, and the unpredictable nature of labour. To alleviate clinicians' instructional burden and enhance trainees' learning efficiency, we introduce a mixed reality (MR) system for childbirth training that combines virtual guidance with tactile manikin interaction, thereby preserving authentic haptic feedback while enabling independent practice without continuous on-site expert supervision. The system extends the passthrough capability of commercial head-mounted displays (HMDs) by spatially calibrating an external RGB-D camera, allowing real-time visual integration of physical training objects. Building on this capability, we implement a coarse-to-fine localization pipeline that first aligns the maternal manikin with fiducial markers to define a delivery region and then registers the pre-scanned neonatal head within this area. This process enables spatially accurate overlay of virtual guiding hands near the manikin, allowing trainees to follow expert trajectories reinforced by haptic interaction. Experimental evaluations demonstrate that the system achieves accurate and stable manikin localization on a standalone headset, ensuring practical deployment without external computing resources. A large-scale user study involving 83 fourth-year medical students was subsequently conducted to compare MR-based and virtual reality (VR)-based childbirth training. Four senior obstetricians independently assessed performance using standardized criteria. Results showed that MR training achieved significantly higher scores in delivery, post-delivery, and overall task performance, and was consistently preferred by trainees over VR training. Although validated only in the context of obstetric delivery, the system demonstrates strong potential for broader manikin-based procedural training and other healthcare education scenarios.

Abstract:
Recovering sewing patterns from draped 3D garments is a challenging problem in human digitization research. Unlike draping from designed sewing pattern with mature physical simulation engines, the inverse process requires mapping complex garment surfaces back to parametric 2D patterns, which existing methods still struggle with. To this end, we propose a two-stage methodology that recovers sewing patterns from 3D garments via an intermediate representation, BoxMesh, which inherently encodes garment-level geometry and panel-level details in 3D space. In Stage I, a geometry-oriented auto-regressive model predicts BoxMesh from the input 3D garment. In Stage II, a semantic-aware auto-regressive model parses BoxMesh into parametric sewing patterns. This decomposition separately tackles geometric inversion and numerical reasoning, enabling more accurate recovery. Experiments have demonstrated that our method not only achieves state-of-the-art performance on the GarmentCodeData benchmark but also can be applied effectively to real scans and single-view images.

Abstract:
Understanding and interpreting a script is essential for effective acting. Existing visualization methods, however, primarily focus on general narrative comprehension and often neglect actors’ specific cognitive and expressive needs. To address this gap, we propose an actor-oriented visual analytics approach that automatically extracts key script features, including implicit emotions, causal relationships, and critical events, to support rapid comprehension of coherent storylines. These extracted features are integrated into a coordinated multi-view visualization, enabling actors to analyze and explore scripts comprehensively and gain deeper insights into the narrative. Furthermore, we introduce a method for structuring performance guidance, converting script events into interpretable, multi-stage recommendations that adapt to individual acting styles. Actor feedback is incorporated to iteratively refine these recommendations, ensuring alignment with personal performance traits. To demonstrate the practicality of our approach, we implement ActorVis, a system that operationalizes these methods and provides actors with interactive visualizations and adaptive performance guidance. Case studies and user evaluations confirm the approach’s effectiveness in facilitating rapid script understanding and enhancing personalized performance support.

Abstract:
This paper presents HaptiCraft, a modular handheld haptic controller designed to replicate the appearance, mass distribution, and multimodal haptic properties of virtual objects. HaptiCraft consists of many modules that provide distinct functions for assembly and multimodal haptic feedback, supporting five types: vibration, impact, thermal, variable inertia, and variable stiffness. Users can readily assemble these modules to create a controller tailored to their needs. To simplify the design process, especially for novice users, we also provide a graphical authoring tool and assess its usability. Finally, we evaluate the system’s effectiveness through various virtual reality (VR) scenarios, demonstrating that HaptiCraft significantly enhances user experience. HaptiCraft makes a substantial contribution to the field of VR handheld controllers by offering comprehensive support for shape-changing, variable inertia, and rich multimodal haptic feedback, coupled with an effective authoring method.

Abstract:
Rendering large-scale, unbounded scenes on AR/VR-class devices is constrained by the computation, bandwidth, and storage cost of 3D Gaussian Splatting (3DGS). We propose a low-power, low-cost 3DGS hardware accelerator that renders full-HD images in real time, together with a hardware-friendly compression pipeline that combines iterative Gaussian pruning and fine-tuning, progressive spherical harmonics (SH) degree reduction, and vector quantization of all SH coefficients and colors. The scheme achieves a 51.6×51.6× model-size reduction with a 0.743 dB PSNR loss. The accelerator uses a frame-level pipeline that integrates point-based culling and projection with tile-based sorting and rasterization, skips zero-Jacobian matrix multiplications (reducing processing elements by 63% and computation by 53%), and adopts comparison-free tile-based sorting with deterministic latency. Implemented in a TSMC 28-nm process at 800MHz, the design occupies \text0.66\;\textmm^20.66mm2 with 1.1438 M gates and 120 kB SRAM, consumes 0.219 W, and delivers 1219 Mpixels/J at 267.5 Mpixels/s, enabling 1080p at 129 FPS. Overall, it is 5.98×5.98× smaller in area, 5.94×5.94× higher throughput, and delivers 7.5×7.5× higher energy efficiency than prior 3DGS accelerators.

Abstract:
In this paper, we introduce DecoRec, a novel system designed to elevate single-view 2D images to a decomposed 3D scene mesh. Current methods for single-view scene reconstruction typically rely on object retrieval or the regression of coarse 3D voxels or surfaces, leading to inaccuracies in capturing the appearance and geometry of the input image. The lack of high-quality large-scale scene-level datasets further complicates direct 3D scene generation from single-view images. To achieve high-quality 3D scene generation from a single-view image, DecoRec takes advantage of recent diffusion-based single-view object reconstruction methods to reconstruct individual objects separately. Subsequently, a refinement pipeline is proposed to effectively merge these reconstructed objects, enhancing appearance and geometry through a differentiable rendering technique and diffusion-guided refinement. Our results demonstrate that DecoRec facilitates high-quality single-view scene reconstruction in both geometry and novel synthesis, offering significant benefits for downstream applications like room interior design.

Abstract:
We present a visualization design study of creating a collaborative virtual reality (VR) system for radiation treatment planning, with an emphasis on proton therapy. The goal is to support teams of dosimetrists, physicians, and medical physicists as they review and compare multiple possible patient-specific treatment plans, which requires analyzing complex 3D spatial relationships between a radiation dosage volume and anatomical structures. The approach is a novel combination and refinement of interactive visualization techniques including: networked multi-user immersive visualization, interactive volume rendering and slicing with 3D widgets and gestures, superimposed surface rendering with GPU-accelerated curvature-directed lines, smart cursors, teleporting, and avatars. These features are integrated within a workflow that supports three complementary modes of visual data comparison (juxtaposition, interchangeable, and explicit encoding). Results and feedback from multi-year iterative development with users and a summative field deployment in the form of a mock plan-review meeting reveal several advantages relative to current clinical practice and suggest directions for future work.

Abstract:
We present GSDeformer, a method that enables cage-based deformation on 3D Gaussian Splatting (3DGS). Our approach bridges cage-based deformation and 3DGS by using a proxy point-cloud representation. This point cloud is generated from 3D Gaussians, and deformations applied to the point cloud are translated into transformations on the 3D Gaussians. To handle potential bending caused by deformation, we incorporate a splitting process to approximate it. Our method does not modify or extend the core architecture of 3D Gaussian Splatting, making it compatible with any trained vanilla 3DGS or its variants. Additionally, we automate cage construction for 3DGS and its variants using a render-and-reconstruct approach. Experiments demonstrate that GSDeformer delivers superior deformation results compared to existing methods, is robust under extreme deformations, requires no retraining for editing, runs in real-time, and can be extended to other 3DGS variants.

Abstract:
We present a real-time pipeline to approximate global illumination in interactive or dynamic scenes, including both 3D Gaussian models and conventional meshes. Building on a formulated surface light transport model for 3D Gaussians, we address key performance challenges through a fast compound stochastic ray-tracing algorithm and a hardware 3D Gaussian rasterizer, and we implement multiple RTGI (real-time global illumination) techniques for 3D Gaussians. Our pipeline enables real-time rendering of interactive scenes featuring editable materials, lights, meshes, and 3D Gaussian models, effectively capturing multi-bounce diffuse and one-bounce glossy light transport. Our approach covers a wide range of dynamic light sources, including area lights, directional lights, and environmental lighting. Comprehensive experimental results demonstrate that our approach can efficiently render global illumination for both 3DGS models and hybrid models combining meshes and 3DGS, all in real time. Our pipeline highlights the potential of 3D Gaussians in real-time global illumination, and offers insights into performance optimization.

Abstract:
Authoring high-quality character animation is essential in multimedia production, especially for pre-rendered formats such as films and TV series. These workflows are often iterative, requiring frequent adjustments and immediate visual feedback, and typically involve editing existing motion, such as mocap data. We present Neural Motion Path (NMP), a deep learning-based system designed to support this authoring process by enabling full-body motion editing through joint-level motion path manipulation. While joint rotations are essential for expressive human motion, explicitly specifying them imposes a significant burden on users; NMP addresses this challenge by enabling intuitive position-only motion path editing while implicitly inferring plausible rotation trajectories. To address the inherent tension between motion plausibility and precise constraint satisfaction, NMP explicitly decouples motion synthesis from constraint enforcement within an autoregressive framework. NMP generates realistic, context-aware motion without fine-tuning every path detail, making it accessible to novices while supporting the demands of detailed animation authoring. It combines a motion generator with a novel RotationNet for inferring joint rotation, and a constraint imposer that enforces end-effector constraints via an Explicit-Weight Sparse Expert Model (EW-SEM). The system supports terrain adaptation and authoring operations like Concatenate, Insert, and Mix. Implemented as a Blender add-on, NMP supports real-time playback and interactive workflows. A user study with novice users shows that NMP improves satisfaction, efficiency, and perceived motion quality compared to the conventional layered keyframing approach, highlighting its potential as an accessible and effective authoring tool.

Abstract:
Video inpainting modifies local regions in video while ensuring spatial and temporal coherence. However, existing methods—both traditional and recent diffusion-based ones—face key limitations: they lack unified support for both insertion and completion, and are restricted to single-object inpainting, making it difficult to handle multi-object scenarios involving grounding and interaction. In this article, we propose MultiPaint, a unified framework for multi-task, multi-object, and multi-condition video inpainting. First, we introduce dual-branch adapters to unify the insertion and completion tasks within a single model. Moreover, we propose a test-time scheduled feature composition strategy that enables multi-object inpainting with user-specified locations while better preserving interactions among objects, a setting that has been insufficiently addressed in prior work. Additionally, we introduce a multi-condition inpainting scheme that integrates text-guided, image-guided, and keyframe-guided modes via dynamic frame masking, providing more controllability in appearance customization. Extensive experiments show that MultiPaint achieves state-of-the-art performance on object insertion and scene completion among the recent works. We further demonstrate its versatility in downstream tasks including grounded video generation, object editing, object removal, image-guided inpainting, and long video inpainting.

Abstract:
Point clouds have become a widely used data format in computer vision, driven by the increasing availability of 3D scanning devices in applications such as robotics and autonomous driving. However, challenges in data acquisition, such as occlusion, reflection, and limited sensor resolution, often result in incomplete scans. Consequently, point cloud completion, the task of reconstructing the complete geometry of a point cloud object from partial data, has emerged as a key area of research. Numerous deep learning-based methods have been proposed to tackle this problem. Although some surveys on shape completion exist, the rapid growth of research in recent years calls for an updated review. In this work, we present a comprehensive survey of recent advancements in deep neural networks for point cloud completion, covering literature from 2024 up to December 2025. After establishing a reference context by analyzing the major foundational methods up to the year 2023, we review the most recent methods with an analytical perspective based on how they have improved each specific component of the network. Furthermore, we present a comparative analysis of the performance of the reviewed methods on the main benchmark datasets, and finally, we discuss open challenges, offering insights and suggestions for future research directions.

Abstract:
Surface offsetting is a fundamental geometric operation in computer-aided design, manufacturing, robotics, and computational physics. Despite its conceptual simplicity, generating offset surfaces robustly and efficiently for complex and irregular geometries remains a persistent challenge, hindered by self-intersections, topological inconsistencies, and feature degradation. This paper reviews recent advances in the field of surface offsetting, offering a structured overview of the evolution of techniques from classical constructive methods to contemporary neural implicit representations. Our survey addresses a gap in the literature, as prior foundational reviews over the past two decades in this field focused primarily on parametric methods. We introduce a taxonomy that organizes existing algorithms into five principal classes: Constructive, Spatial Discretization, Optimization-based, Field-based, and Learning-based approaches. An analysis of 46 representative algorithms reveals trade-offs in algorithm design: achieving both geometric accuracy and topological correctness proves difficult, and representation choices introduce inherent complexity constraints. We identify open problems including offset generation for surfaces with open boundaries, preservation of thin features, and resolution of self-intersections in concave regions. These challenges point toward promising research directions, including extending classical offset theory to non-manifold and open-boundary domains, developing scalable geometric predicates, and designing hybrid neural representations that disentangle distance-field smoothness from geometric sharpness.

Abstract:
Volumetric video and 6-DoF scene capture are becoming central to immersive applications such as telepresence and mixed reality content delivery. However, existing volumetric datasets are often short in duration, restricted to studio-captured human subjects, and provide only limited geometric representations. Consequently, evaluating real-world immersive applications in full-scene contexts often necessitates custom capture and 3D reconstruction setups, creating high practical barriers and ultimately hindering reproducibility. To this end, we present SceneHub4D, a new dataset and evaluation framework. Our dataset captures long, dynamic sequences across diverse real-world indoor environments with synchronized multi-view RGB-D streams, calibrated camera poses, and high-resolution background geometry reconstructed via photogrammetry and LiDAR. We provide multiple 3D representations, including point clouds, textured meshes, and Gaussian splats, along with a software toolkit for format conversion, rendering, and metric evaluation. To support structured comparison and perceptual analysis, we provide supplementary metrics including Geometry Complexity Score and Volumetric Temporal Information, and evaluate rendering performance across desktop GPUs and VR headsets. By lowering the practical barriers to capture, reconstruction, and evaluation, SceneHub4D enables researchers to study immersive 3D streaming and rendering systems without requiring custom hardware setups or complex data collection pipelines. We expect it will serve as a useful foundation for advancing volumetric media research.

Abstract:
Extended Reality (XR) technologies offer innovative opportunities to address inequities in laboratory preparation among first-year undergraduate chemistry students, particularly benefiting students from under-served communities. This study presents the development and evaluation of an immersive XR module specifically designed to teach glove hygiene, a critical laboratory safety skill frequently overlooked in traditional high school curricula due to resource constraints. Involving 176 student test cases across diverse institutions, this module integrates active learning and inclusive, equity-driven design principles. A previously piloted VR glove hygiene module revealed significant student enthusiasm but encountered challenges related to nausea during immersive experiences. Addressing this, our refined XR module employs practices for nausea reduction, providing an accessible virtual lab environment with interactive glove hygiene simulations. Preliminary evaluations indicate substantial improvements in students' confidence, reduction in laboratory-related anxiety, and increased engagement, highlighting XR's potential to close preparation gaps. This work underscores the broader impact of XR interventions in STEM education, fostering greater retention and success for underrepresented and first-generation college students.

Abstract:
Social Virtual Reality (VR) platforms provide immersive social experiences but also expose users to serious risks of online harassment. Existing safety measures are largely reactive, while proactive solutions that detect harassment behavior during an incident often depend on sensitive biometric data, raising privacy concerns. In this paper, we present HarassGuard, a vision-language model (VLM) based system that detects physical harassment in social VR using only visual input. We construct an IRB-approved harassment vision dataset, apply prompt engineering, and fine-tune VLMs to detect harassment behavior by considering contextual information in social VR. Experimental results demonstrate that HarassGuard achieves competitive performance compared to state-of-the-art baselines (i.e., LSTM/CNN, Transformer), reaching an accuracy of up to 88.09% in binary classification and 68.85% in multi-class classification. Notably, HarassGuard matches these baselines while using significantly fewer fine-tuning samples (200 vs. 1,115), offering unique advantages in contextual reasoning and privacy-preserving detection.

Abstract:
Creating interactive VR content typically requires extensive modeling effort, making scalable production difficult. We present an automated computational pipeline that converts a single image into an interactive 3D scene with plausible kinesthetic feedback. Our approach leverages a Large Language Model (LLM) to detect objects in the image and infer haptic properties from visual and textual cues. These extracted results are then utilized to synthesize 3D models and optimize haptic properties for perceptual distinguishability, while combining them into a complete interactive environment. A user study shows that the generated VR scenes provide compelling visuo-haptic experiences, highlighting the potential of our method for scalable multisensory world generation. To our knowledge, this is the first system to automatically produce VR scenes with force feedback from a single image, pointing toward a practical direction for lowering the barriers to creating haptically enriched VR content.

Abstract:
Searching for information, objects, or places in virtual reality (VR) is often a cumbersome process that breaks user immersion. Existing interaction techniques like pointing or voice commands are context-dependent, yet current systems fail to help users select the optimal strategy for a given situation. To address this challenge, we introduce an adaptive VR search framework that recommends search strategies based on situational factors. We first propose a taxonomy of five common VR search strategies and construct a factorial dataset linking task types, object properties, and environmental conditions to strategy choices. We then conduct a two-stage study. In Stage 1, we capture users' natural strategy selections in a free-search environment to create a ground-truth dataset. In Stage 2, we use this data to train and evaluate two adaptive systems, one driven by a machine learning (ML) model and another by a Large Language Model (LLM), against the free-search baseline. Our results show that both adaptive systems significantly reduce the number of attempts and perceived workload compared to free search. While the ML-based system achieved the fastest task completion times, both adaptive approaches were rated as significantly more usable. This work demonstrates that adaptive, context-aware systems can enhance search efficiency and user experience in VR, paving the way for more intelligent and immersive information-seeking interfaces.

Abstract:
The transition from conceptual sketch to 3D model poses a major challenge in engineering design. While Mixed Reality (MR) enables direct 3D creation, designers face two key challenges: the Initiation Challenge (lack of meaningful starting points) and the Expression Challenge (difficulty in precise mid-air sketching). We introduce GenerativeSurface, leveraging AI to translate natural language descriptions into interactive 3D surfaces that aid designers in 3D sketching, simultaneously addressing both challenges. Our system transforms the design workflow into a fluid “describe, generate, sketch, iterate” cycle. Comparative evaluation against primitive scaffolding and freehand mid-air sketching demonstrates that our approach effectively bridges the gap between designers' conceptual intent and 3D sketch expression in immersive environments.

Abstract:
Effective Augmented Reality (AR) guidance for complex assembly faces a dual challenge: the inability of conventional liaison graphs to represent procedural logic, and the cognitive burden imposed by visual instructions. We argue that the solution requires a more expressive structure to overcome these representational deficits and a narration approach to mediate instruction complexity. Our method first employs an assembly hypergraph to capture the task's hierarchical information, from which an A search algorithm generates an optimal assembly path. Then a Large Language Model (LLM)-mediated narration workflow is designed to address the ergonomic deficiencies of the machine-centric path. It employs an optimizer to improve fluency, followed by a narrator that crafts the steps into an intuitive instruction narration. A within-subjects user study (N = 24) revealed a progressive enhancement from our method's components. The transition from a liaison-graph baseline to the hypergraph alone improved objective outcomes by reducing task time and errors and improving subjective ratings (SUS, NASA-TLX, TAM, and ARI). Subsequently, augmenting the LLM-mediated narration maintained these gains while lowering cognitive load and elevating user experience and usability. Our findings indicate the value of our AR assembly design and discuss the opportunities of using LLM as a mediation layer for better user interaction.

Abstract:
Virtual reality is becoming increasingly popular, and modern haptic equipment, such as vibrotactile suits, haptic gloves, and force-feedback controllers, offers new means of interaction within virtual environments, significantly enhancing user experience. When interacting with virtual objects, combined visual and haptic feedback simulates the physical sensations of grasping, lifting, or moving real objects. This sensorimotor feedback is essential for inducing a sense of presence and agency, yet it remains challenging to reproduce in the absence of reliable haptic cues. In this study, we design and evaluate several haptic metaphors using combinations of vibrotactile design parameters to simulate the lifting effort associated with light to heavy objects. These parameters include primitive signals, intensity, spatial density, propagation, and temporal density. Our contribution is threefold. First, we propose a method for modulating perceived physical effort by extending signal intensity with spatial and temporal density, which together reflect the effort required to lift an object. Second, we present a user study in which participants compared haptic effects and ranked them according to perceived lifting effort, comfort, and confidence, allowing us to assess the influence of each parameter. Third, we report the results of a second study in which participants evaluated vibrotactile effects when lifting different virtual objects. The findings confirm the importance of intensity and spatial density, as well as the influence of graphical representation on perceived effort. This research provides practical insights for designing haptic-enabled virtual reality systems and offers guidance for developers seeking to create more expressive and believable vibrotactile interactions.

Abstract:
This paper presents an interactive analysis framework for exploring data from photoelastic disk experiments, which serve as a model for two-dimensional granular materials. Granular materials, composed of discrete particles such as sand or gravel, exhibit behaviors resembling fluid or solid states depending on the system configuration. These behaviors arise from interparticle contact forces, which form complex force networks that govern the material’s macroscopic behavior. Our framework is specifically designed to analyze such 2D ensembles of dynamic force networks, enabling the identification and characterization of their underlying structures. The framework is built around a topology-based, multiscale data segmentation in terms of force chains and cycles. The analysis methods are structured across three levels: (1) multiscale analysis of individual instances under specific loading conditions, (2) detailed exploration of single experiments encompassing a series of loading and unloading cycles, and (3) comparative analysis across experiments conducted under similar and differing setups. We demonstrate the capabilities of our framework with a case study for each of these levels.

Abstract:
Generating high-fidelity, expressive, and realistic 3D head avatars remains a fundamental challenge for immersive applications such as virtual reality, gaming, and telepresence. This task requires not only precise modeling of non-rigid facial deformations but also semantically controllable expression synthesis under diverse viewpoints and motion contexts. We present DynAvatar, a novel framework that integrates expression-guided deformation into the 3D Gaussian splatting pipeline to produce photorealistic and emotionally resonant head avatars. Our method introduces two key innovations: (1) an expression-guided Gaussian deformation module that tightly couples geometric displacement with high-level semantic cues, enabling fine-grained and anatomically meaningful facial animation; and (2) a spatial context embedding mechanism that encodes the canonical position of each Gaussian to preserve semantic coherence and spatial consistency during expression generation. Extensive experiments on both controlled and in-the-wild datasets demonstrate that DynAvatar significantly outperforms state-of-the-art methods in terms of visual realism, expression fidelity, and rendering quality.

Abstract:
Contemporary materials science research is heavily conducted in silico, involving massive simulations of the atomic-scale evolution of materials. Cataloging basic patterns in the atomic displacements is key to understanding and predicting the evolution of physical properties. However, the combinatorial complexity of the space of possible transitions coupled with the overwhelming amount of data being produced by high-throughput simulations make such an analysis extremely challenging and time-consuming for domain experts. The development of visual analytics systems that facilitate the exploration of simulation data is an active field of research. While these systems excel in identifying temporal regions of interest, they treat each timestep of a simulation as an independent event without considering the behavior of the atomic displacements between timesteps. We address this gap by introducing LAMDA, a visual analytics system that allows domain experts to quickly and systematically explore state-to-state transitions. In LAMDA, transitions are hierarchically categorized, providing a basis for cataloging displacement behavior, as well as enabling the analysis of simulations at different resolutions, ranging from very broad qualitative classes of transitions to very narrow definitions of unit processes. LAMDA supports navigating the hierarchy of transitions, enabling scientists to visualize the commonalities between different transitions in each class in terms of invariant features characterizing local atomic environments, and LAMDA simplifies the analysis by capturing user inputs through annotations. We evaluate our system through a case study and report on findings from our domain experts.

Abstract:
Most current audio-driven facial animation research primarily focuses on generating videos with neutral emotions. While some studies have addressed the generation of facial videos driven by emotional audio, efficiently generating high-quality talking head videos that integrate both emotional expressions and style features remains a significant challenge. In this paper, we propose ESGaussianFace, an innovative framework for emotional and stylized audio-driven facial animation. Our approach leverages 3D Gaussian Splatting to reconstruct 3D scenes and render videos, ensuring efficient generation of 3D consistent results. We propose an emotion-audio-guided spatial attention method that effectively integrates emotion features with audio content features. Through emotion-guided attention, the model is able to reconstruct facial details across different emotional states more accurately. To achieve emotional and stylized deformations of the 3D Gaussian points through emotion and style features, we introduce two 3D Gaussian deformation predictors. Futhermore, we propose a multi-stage training strategy, enabling the step-by-step learning of the character’s lip movements, emotional variations, and style features. Our generated results exhibit high efficiency, high quality, and 3D consistency. Extensive experimental results demonstrate that our method outperforms existing state-of-the-art techniques in terms of lip movement accuracy, expression variation, and style feature expressiveness.

Abstract:
Understanding embedding relationships is crucial for neural network interpretability, structural analysis, and data exploration. However, visually comparing embeddings is challenging due to the difficulty of mentally aligning structures across views. In this paper, we address this challenge by constructing multiple hierarchies of data points from different perspectives to facilitate meaningful comparisons. We introduce Parallel Clusters (ParaClus), a visual analytics system that enables multi-scale exploration of embedding structures. This is achieved through cluster formations across embeddings and attributes, along with an adaptive thresholding mechanism that defines neighborhoods. By dynamically adjusting this threshold, users can explore structures at varying scales. Our interface employs a parallel axes design, where clusters derived from the same embedding or attribute are aligned along individual axes. This layout allows users to easily compare neighboring clusters and observe relationships across embeddings. Furthermore, interactive subdivision mechanisms enable users to refine clusters based on their connections to other clusters, providing deeper insights into structural dependencies. Additionally, our system seamlessly integrates labels and scalar attributes into the clustering process, offering a unified approach to analyzing multi-attribute, time-varying, and network-derived embeddings. We evaluate the effectiveness of ParaClus through expert assessments and case studies.

Abstract:
Multivariate data contain an abundance of information and many techniques have been proposed to allow humans to navigate this information in an ordered fashion. For this work, we focus on methods that seek to convey multivariate data as a collection of bivariate scatterplots or parallel coordinates plots. Presenting multivariate data in this way requires a regime that determines in what order the bivariate scatterplots are presented or in what order the parallel coordinate axes are arranged. We refer to this order as a visualization sequence. Common techniques utilize standard statistical metrics like correlation, similarity or consistency. We expand on the family of statistical metrics by incorporating the rigidity of causal relationships. To capture these relationships, we first derive a causal graph from the data and then allow users to select from several semantic traversal schemes to derive the respective chart sequence. We tested the sequences with a crowd-sourced user study and a user interview to confirm that the causality-informed visualization sequences help viewers to better grasp the causal relationships that exist in the data, as opposed to sequences derived from correlations or randomization alone.

Abstract:
Generating full-body humans in 360^\circ ∘ has broad applications in digital entertainment, online education and art design. Existing works primarily rely on coarse conditions such as body pose to guide the generation, lacking detailed control over the synthesized results. Regarding this limitation, sketches offer a promising alternative as an expressive condition that enables more explicit and precise control. However, current sketch-based generation methods focus on faces or common objects, how to transfer sketches into 360 ^\circ ∘ full-body humans remains unexplored. To bridge this gap, we propose Sketch2Avatar, the first generative model to achieve 3D full-body human generation from hand-drawn sketches. Our model is capable of synthesizing sketch-aligned and 360^\circ ∘-consistent full-body human images by leveraging the geometry information extracted from sketches to guide the 3D representation generation and neural rendering. Specifically, we propose sketchguided 3D representation generation to model the 3D human and maintain the alignment between input sketches and generated humans. Our transformer-based generator incorporates spatial feature guidance and latent modulation derived from sketches to produce high-quality 3D representations. Additionally, our designed bodyaware neural rendering utilizes 3D human body priors from sketches, simplifying the learning of articulated body poses and complex body shapes. To train and evaluate our model, we construct a large-scale dataset comprising approximately 19 K 2D full-body human images and their corresponding sketches in a hand-drawn style. Experimental results demonstrate that our Sketch2Avatar can transfer hand-drawn sketches into photo-realistic 360^\circ ∘ full-body human images with precise sketch-human alignment. Ablation studies further validate the effectiveness of our design choices.

Abstract:
Reconstructing 3D scenes and synthesizing novel views from sparse input views is a highly challenging task. Recent advances in video diffusion models have demonstrated strong temporal reasoning capabilities, making them a promising tool for enhancing reconstruction quality under sparse-view settings. However, existing approaches are primarily designed for modest viewpoint variations, which struggle in capturing fine-grained details in close-up scenarios since input information is severely limited. In this paper, we present a diffusion-based framework, called CloseUpShot, for close-up novel view synthesis from sparse inputs via point-conditioned video diffusion. Specifically, we observe that pixel-warping conditioning suffers from severe sparsity and background leakage in close-up settings. To address this, we propose hierarchical warping and occlusion-aware noise suppression, enhancing the quality and completeness of the conditioning images for the video diffusion model. Furthermore, we introduce global structure guidance, which leverages a dense fused point cloud to provide consistent geometric context to the diffusion process, to compensate for the lack of globally consistent 3D constraints in sparse conditioning inputs. Extensive experiments on multiple datasets demonstrate that our method outperforms existing approaches, especially in close-up novel view synthesis, clearly validating the effectiveness of our design.

Abstract:
2D Gaussian Splatting has recently emerged as a significant method in 3D reconstruction, enabling novel view synthesis and geometry reconstruction simultaneously. While the well-known Gaussian kernel is broadly used, its lack of anisotropy and deformation ability leads to dim and vague edges at object silhouettes, limiting the reconstruction quality of current Gaussian splatting methods. To enhance the representation power, we draw inspiration from quantum physics and propose to use the Gaussian-Hermite kernel as the new primitive in Gaussian splatting. The new kernel takes a unified mathematical form and extends the Gaussian function, which serves as the zero-rank special case in the updated general formulation. Our experiments demonstrate that the proposed Gaussian-Hermite kernel achieves improved performance over traditional Gaussian Splatting kernels on both geometry reconstruction and novel-view synthesis tasks. Specifically, on the DTU dataset, our method yields more accurate geometry reconstruction, while on datasets such as MipNeRF360 and our customized Detail dataset, it achieves better results in novel-view synthesis. These results highlight the potential of the Gaussian-Hermite kernel for high-quality 3D reconstruction and rendering.

Abstract:
Mixed Reality Agents (MiRAs) have been extensively studied to enhance virtual-physical interactions, using their ability to exist in both virtual and physical environments. However, little research has focused on enhancing perceived empathy in MiRAs, despite its potential for agent-assisted therapy, education, and training. To fill this gap, we investigate the impact of an Empathic Mixed Reality agent (EMiRA) that adapts to users’ physiological states and physical events in a shooting game. We found that this adaptation enhanced users’ social perceptions of the agent, including social presence, social connectedness, and perceived empathy. Physiological adaptation increased paternalism and reduced user dominance, while physical adaptation had no such effect. We discuss these findings and provide design implications for future EMiRAs.

Abstract:
Reconstructing Dynamic 3D Gaussian Splatting (3DGS) from low-framerate RGB videos is challenging. This is because large inter-frame motions will increase the uncertainty of the solution space. For example, one pixel in the first frame might have more choices to reach the corresponding pixel in the second frame. Event cameras can asynchronously capture rapid visual changes and are robust to motion blur, but they do not provide color information. Intuitively, the event stream can provide deterministic constraints for the inter-frame large motion by the event trajectories. Hence, combining low-temporal resolution images with high-framerate event streams can address this challenge. However, it is challenging to jointly optimize Dynamic 3DGS using both RGB and event modalities due to the significant discrepancy between these two data modalities. This paper introduces a novel framework that jointly optimizes dynamic 3DGS from the two modalities. The key idea is to adopt event motion priors to guide the optimization of the deformation fields. First, we extract the motion priors encoded in event streams by using the proposed LoCM unsupervised fine-tuning framework to adapt an event flow estimator to a certain unseen scene. Then, we present the geometry-aware data association method to build the event-Gaussian motion correspondence, which is the primary foundation of the pipeline, accompanied by two useful strategies, namely motion decomposition and inter-frame pseudo-label. Extensive experiments show that our method outperforms existing image and event-based approaches across synthetic and real scenes and prove that our method can effectively optimize dynamic 3DGS with the help of event data.

Abstract:
Graph Neural Networks (GNNs) have achieved significant success across various applications. However, their complex structures and inner workings can be challenging for non-AI experts to understand. To address this issue, this study presents GNN101, an educational visualization tool for interactive learning of GNNs. GNN101 introduces a set of animated visualizations that seamlessly integrates mathematical formulas with visualizations via multiple levels of abstraction, including a model overview, layer operations, and detailed calculations. Users can easily switch between two complementary views: a node-link view that offers an intuitive understanding of the graph data, and a matrix view that provides a space-efficient and comprehensive overview of all features and their transformations across layers. GNN101 was designed and developed based on close collaboration with four GNN experts and deployment in three GNN-related courses. We demonstrated the usability and effectiveness of GNN101 via use cases and user studies with both GNN teaching assistants and students. To ensure broad educational access, GNN101 is developed through modern web technologies and available directly in web browsers without requiring any installations.

Abstract:
Scatterplots are widely used across various domains to identify anomalies in datasets, particularly in multi-class settings, such as detecting misclassified or mislabeled data. However, scatterplot effectiveness often declines with large datasets due to limited display resolution. This paper introduces a novel Visual Quality Measure (VQM) – OM4AnI (Overlap Measure for Anomaly Identification) – which quantifies the degree of overlap for identifying anomalies, helping users estimate how effectively anomalies can be observed in multi-class scatterplots. OM4AnI begins by computing anomaly index based on each data point’s position relative to its class cluster. The scatterplot is then discretized into a matrix representation by binning the display space into cell-level (pixel-level) grids and computing the coverage for each pixel. It takes into account the anomaly index of data points covering these pixels and visual features (marker shapes, marker sizes, and rendering orders). Building on this foundation, we sum all the coverage information in each cell (pixel) of matrix representation to obtain the final quality score with respect to anomaly identification. We conducted an evaluation to analyze the efficiency, effectiveness, sensitivity of OM4AnI in comparison with six representative baseline methods that are based on different computation granularity levels: data level, marker level, and pixel level. The results show that OM4AnI outperforms baseline methods by exhibiting more monotonic trends against the ground truth and greater sensitivity to rendering order, unlike the baseline methods. It confirms that OM4AnI can inform users about how effectively their scatterplots support anomaly identification. Overall, OM4AnI shows strong potential as an evaluation metric and for optimizing scatterplots through automatic adjustment of visual parameters.

Abstract:
Effective visual accessibility in Virtual Reality (VR) is crucial for Blind and Low Vision (BLV) users. However, designing visual accessibility systems is challenging due to the complexity of 3D VR environments and the need for techniques that can be easily retrofitted into existing applications. While prior work has studied how to enhance or translate visual information, the advancement of Vision Language Models (VLMs) provides an exciting opportunity to advance the scene interpretation capability of current systems. This paper presents EnVisionVR, an accessibility tool for VR scene interpretation. Through a formative study of usability barriers, we confirmed the lack of visual accessibility features as a key barrier for BLV users of VR content and applications. In response, we used our findings from the formative study to inform the design and development of EnVisionVR, a novel visual accessibility system leveraging a VLM, voice input and multimodal feedback for scene interpretation and virtual object interaction in VR. An evaluation with 12 BLV users demonstrated that EnVisionVR significantly improved their ability to locate virtual objects, effectively supporting scene understanding and object interaction.

Abstract:
Recent advances in dynamic garment reconstruction boost virtual try-on with monocular video streams as inputs. However, existing literature has been intensively conducted on its sub-tasks, including static garment reconstruction and dynamic clothed human reconstruction, which are difficult to extend to dynamic garment reconstruction. The former bottleneck is mainly the lack of cross-frame correspondences and independent clothing topology on implicit garment fields, which results in the inability to obtain accurate motion information during dynamic clothing reconstruction and the absence of stable topology critical for downstream tasks, such as animation with physics engines. The latter usually binds the garment motion with body or skeleton motions, leading to rigid artifacts for loose-fitting garments. Our key idea is to build a diffusion based T-pose garments generator with a strong prior on garments structure. The garment generator is trained to generate 2D clothing representation, termed FOSUP, conditioned by a monocular video. FOSUP, defined as FOurier Spherical Unwrapping, enables a bidirectional mapping between FOSUP and the mesh through FFT and inverse FFT, which maintain spatial order and adjacency. Subsequently, this FOSUP is mapped back to 3D meshes through an inverse FFT process and transformed into pose space through a point transformation network to guide the three-dimensional reconstruction of the entire sequence. To sufficiently train our framework and address the lack of domain-specific data, we have constructed a large-scale garment MoCap dataset. This dataset captures the motion of various loose garments and includes multi-view raw images, frame-by-frame human motion annotations, raw scanned point clouds, topology-independent garment templates, and garment meshes with cross-frame correspondences. Comprehensive experiments have demonstrated that our unwrapping-based representation and diffusion-based framework significantly improve the performance and robustness of dynamic garment reconstruction.

Abstract:
Visualization design is often described as a process of solving a well-defined problem by navigating a design space. While existing visualization design models have provided valuable structure and guidance, they tend to foreground technical problem-solving and underemphasize the interpretive, judgment-based aspects of design. In contrast, research in other design disciplines has emphasized the importance of framing-how designers define and redefine what the problem is-and the co-evolution of problem and solution spaces through reflective practice. These dimensions remain underexplored in visualization research, particularly from the perspective of expert practitioners. This paper investigates how visualization designers frame problems and navigate the interplay between problem understanding and solution development. We conducted a mixed-methods study with 11 expert design practitioners using design challenges, diary entries, and semi-structured interviews. Through reflexive thematic analysis, we identified key strategies that participants used to frame design problems, reframe them in response to evolving constraints or insights, and construct bridges between problem and solution spaces. These included the use of metaphors, heuristics, sketching, primary generators, and reflective evaluation of failed or incomplete ideas. Our findings contribute an empirically grounded account of visualization design as a reflective, co-evolutionary practice. We show that framing is not a preliminary step, but a continuous activity embedded in the act of designing. Participants frequently shifted their understanding of the problem based on solution attempts, feedback from tools, and ethical or narrative concerns. These insights extend current visualization design models and highlight the need for frameworks that better account for framing and interpretive judgment. We conclude with implications for visualization research, education, and practice. In particular, we discuss how design education can better support framing and co-evolutionary thinking, and how visualization research can benefit from greater attention to the cognitive strategies and reflective processes that underpin expert design.

Abstract:
Open-world object detection (OWOD) extends traditional object detection to identifying both known and unknown object, necessitating continuous model adaptation as new annotations emerge. Current approaches face significant limitations: 1) data-hungry training due to reliance on a large number of crowdsourced annotations, 2) susceptibility to “partial feature overfitting,” and 3) limited flexibility due to required model architecture modifications. To tackle these issues, we present OW-CLIP, a visual analytics system that provides curated data and enables data-efficient OWOD model incremental training. OW-CLIP implements plug-and-play multimodal prompt tuning tailored for OWOD settings and introduces a novel “Crop-Smoothing” technique to mitigate partial feature overfitting. To meet the data requirements for the training methodology, we propose dual-modal data refinement methods that leverage large language models and cross-modal similarity for data generation and filtering. Simultaneously, we develope a visualization interface that enables users to explore and deliver high-quality annotations—including class-specific visual feature phrases and fine-grained differentiated images. Quantitative evaluation demonstrates that OW-CLIP achieves competitive performance at 89% of state-of-the-art performance while requiring only 3.8% self-generated data, while outperforming SOTA approach when trained with equivalent data volumes. A case study shows the effectiveness of the developed method and the improved annotation quality of our visualization system.

Abstract:
We present a data-domain sampling regime for quantifying CNNs' graphic perception behaviors. This regime lets us evaluate CNNs' ratio estimation ability in bar charts from three perspectives: sensitivity to training-test distribution discrepancies, stability to limited samples, and relative expertise to human observers. After analyzing 16 million trials from 800 CNN models and 6,825 trials from 113 human participants, we arrived at a simple and actionable conclusion: CNNs can outperform humans and their biases simply depend on the training-test distance. We show evidence of this simple, elegant behavior of the machines when they interpret visualization images. osf.io/gfqc3 provides registration, the code for our sampling regime, and experimental results.

Abstract:
Implicit neural representations (INRs) have emerged as a transformative paradigm for time-varying volumetric data compression and representation, owing to their ability to model high-dimensional signals effectively. INRs represent scalar fields based on sampled coordinates, typically using either a single network for the entire field or multiple networks across different spatial domains. However, these approaches often face challenges in modeling complex patterns and introducing boundary artifacts. To address these limitations, we propose MoE-INR, an INR architecture based on a mixture-of-experts (MoE) framework. MoE-INR automates irregular subdivisions of spatiotemporal fields and dynamically assigns them to different expert networks. The architecture comprises three key components: a policy network, a shared encoder, and multiple expert decoders. The policy network subdivides the field and determines which expert decoder is responsible for a given input coordinate. The shared encoder extracts hidden representations from the input coordinates, and the expert decoders transform these high-dimensional features into scalar values. This design results in a unified framework accommodating diverse INR types, including conventional, grid-based, and ensemble. We evaluate the effectiveness of MoE-INR on multiple time-varying datasets with varying characteristics. Experimental results demonstrate that MoE-INR significantly outperforms existing non-MoE and MoE-based INRs and traditional lossy compression methods across quantitative and qualitative metrics under various compression ratios.

Abstract:
Design studies aim to develop visualization solutions for real-world problems across various application domains. Recently, the emergence of large language models (LLMs) has introduced new opportunities to enhance the design study process, providing capabilities such as creative problem-solving, data handling, and insightful analysis. However, despite their growing popularity, there remains a lack of systematic understanding of how LLMs can effectively assist researchers in visualization-specific design studies. In this paper, we conducted a rnulti-stage qualitative study to fill this gap, which involved 30 design study researchers from diverse backgrounds and expertise levels. Through in-depth interviews and carefully-designed questionnaires, we investigated strategies for utilizing LLMs, the challenges encountered, and the practices used to overcome them. We further compiled the roles that LLMs can play across different stages of the design study process. Our findings highlight practical implications to inform visualization practitioners, and also provide a framework for leveraging LLMs to facilitate the design study process in visualization research.

Abstract:
Linear embeddings support interactive visual exploration by mapping high-dimensional (nD) data into a two-dimensional space. Despite their popularity, selecting meaningful projection parameters remains a key challenge due to the infinite 2n-dimensional parameter space. Once an informative projection is found, users often seek similar ones that emphasize specific items differently while preserving global structure. For instance: Do clusters become outliers under slight changes? Can grouped items separate-or merge-through parameter adjustments? Which changes to the embedding parameters lead to such projections - and do they exist at all? Answering these questions efficiently is critical for effective visual search. Yet, current methods-such as projection tours or manual parameter tuning-are time-consuming and risk overlooking important views, including those of specific interest. We propose Composition Operators, a mathematical foundation for a novel set-of-point manipulation concept for linear embeddings-such as Star Coordinates-as an alternative approach to selecting informative embedding parameters in a more controllable manner with respect to the desired outcome. Users specify item-based constraints on the projection result; the corresponding 2n parameters are then derived automatically, eliminating the need to exhaustively search the entire parameter space to get a similar outcome. Neither the embedding space nor the set of parameters is altered - only the mechanism for navigating and selecting parameters is redefined. We provide closed-form solutions for this and demonstrate our interactive prototype on nD datasets from the UCI repository.

Abstract:
Large-scale scientific simulations require significant resources to generate high-resolution time-varying data (TVD). While super-resolution is an efficient post-processing strategy to reduce costs, existing methods rely on a large amount of HR training data, limiting their applicability to diverse simulation scenarios. To address this constraint, we proposed CD-TVD, a novel framework that combines contrastive learning and an improved diffusion-based super-resolution model to achieve accurate 3D super-resolution from limited time-step high-resolution data. During pre-training on historical simulation data, the contrastive encoder and diffusion super-resolution modules learn degradation patterns and detailed features of high-resolution and low-resolution samples. In the training phase, the improved diffusion model with a local attention mechanism is fine-tuned using only one newly generated high-resolution timestep, leveraging the degradation knowledge learned by the encoder. This design minimizes the reliance on large-scale high-resolution datasets while maintaining the capability to recover fine-grained details. Experimental results on fluid and atmospheric simulation datasets confirm that CD-TVD delivers accurate and resource-efficient 3D super-resolution, marking a significant advancement in data augmentation for large-scale scientific simulations. The code is available at https://github.com/Xin-Gao-private/CD-TVD.

Abstract:
In this paper, we present a novel compression framework, TFZ, that preserves the topology of 2D symmetric and asymmetric second-order tensor fields defined on flat triangular meshes. A tensor field assigns a tensor—a multi-dimensional array of numbers—to each point in space. Tensor fields, such as the stress and strain tensors, and the Riemann curvature tensor, are essential to both science and engineering. The topology of tensor fields captures the core structure of data, and is useful in various disciplines, such as graphics (for manipulating shapes and textures) and neuroscience (for analyzing brain structures from diffusion MRI). Lossy data compression may distort the topology of tensor fields, thus hindering downstream analysis and visualization tasks. TFZ ensures that certain topological features are preserved during lossy compression. Specifically, TFZ preserves degenerate points essential to the topology of symmetric tensor fields and retains eigenvector and eigenvalue graphs that represent the topology of asymmetric tensor fields. TFZ scans through each cell, preserving the local topology of each cell, and thereby ensuring certain global topological guarantees. We showcase the effectiveness of our framework in enhancing the lossy scientific data compressors SZ3 and SPERR.

Abstract:
Detecting and interpreting common patterns in relational data is crucial for understanding complex topological structures across various domains. These patterns, or network motifs, can often be detected algorithmically. However, visual inspection remains vital for exploring and discovering patterns. This paper focuses on presenting motifs within BioFabric network visualizations—a unique technique that opens opportunities for research on scaling to larger networks, design variations, and layout algorithms to better expose motifs. Our goal is to show how highlighting motifs can assist users in identifying and interpreting patterns in BioFabric visualizations. To this end, we leverage existing motif simplification techniques. We replace edges with glyphs representing fundamental motifs such as staircases, cliques, paths, and connector nodes. The results of our controlled experiment and usage scenarios demonstrate that motif simplification for BioFabric is useful for detecting and interpreting network patterns. Our participants were faster and more confident using the simplified view without sacrificing accuracy. The efficacy of our current motif simplification approach depends on which extant layout algorithm is used. We hope our promising findings on user performance will motivate future research on layout algorithms tailored to maximizing motif presentation. Our supplemental material is available at https://osf.io/f8s3g/?view_only=7e2df9109dfd4e6c85b89ed828320843.

Abstract:
In sports analytics, tactical visualization is widely used to convey valuable insights. However, due to the complex domain knowledge and contextual information involved in tactical visualizations, it is challenging for users to connect high-level tactical insights to corresponding visual patterns. This requires users to engage in a reasoning process to interpret insights within game contexts, which remains insufficiently supported in existing visual-text linking studies. In this work, we propose InsightChaser, a novel approach to bridge tactical insights and soccer visualizations through visual-text linking and visual reasoning enhancement. InsightChaser constructs knowledge graphs to represent both visual elements and contextual game information. Integrating large language models (LLMs), our approach retrieves relevant visual elements and establishes explicit links with insights. Moreover, InsightChaser utilizes LLMs to enhance these visual-text links by providing reasoning explanations and visual effects. We further develop an interactive visualization system that supports navigation and explanation of enhanced visual-text links. Users can explore linked tactical insights interactively and reason through enhanced visual explanations. We conduct two case studies using real-world soccer data and a user study to demonstrate the effectiveness of our approach.

Abstract:
Creating aesthetically pleasing data visualizations remains challenging for users without design expertise or familiarity with visualization tools. To address this gap, we present DataWink, a system that enables users to create custom visualizations by adapting high-quality examples. Our approach combines large multimodal models (LMMs) to extract data encoding from existing SVG-based visualization examples, featuring an intermediate representation of visualizations that bridges primitive SVG and visualization programs. Users may express adaptation goals to a conversational agent and control the visual appearance through widgets generated on demand. With an interactive interface, users can modify both data mappings and visual design elements while maintaining the original visualization's aesthetic quality. To evaluate DataWink, we conduct a user study (N=12) with replication and free-form exploration tasks. As a result, DataWink is recognized for its learnability and effectiveness in personalized authoring tasks. Our results demonstrate the potential of example-driven approaches for democratizing visualization creation.

Abstract:
Understanding the behavior of large language models (LLMs) is crucial for ensuring their safe and reliable use. However, existing explainable AI (XAI) methods for LLMs primarily rely on word-level explanations, which are often computationally inefficient and misaligned with human reasoning processes. Moreover, these methods often treat explanation as a one-time output, overlooking its inherently interactive and iterative nature. In this paper, we present LLM Analyzer, an interactive visualization system that addresses these limitations by enabling intuitive and efficient exploration of LLM behaviors through counterfactual analysis. Our system features a novel algorithm that generates fluent and semantically meaningful counterfactuals via targeted removal and replacement operations at user-defined levels of granularity. These counterfactuals are used to compute feature attribution scores, which are then integrated with concrete examples in a table-based visualization, supporting dynamic analysis of model behavior. A user study with LLM practitioners and interviews with experts demonstrate the system's usability and effectiveness, emphasizing the importance of involving humans in the explanation process as active participants rather than passive recipients.

Abstract:
Exploring volumetric data is crucial for interpreting scientific datasets. However, selecting optimal viewpoints for effective navigation can be challenging, particularly for users without extensive domain expertise or familiarity with 3D navigation. In this paper, we propose a novel framework that leverages natural language interaction to enhance volumetric data exploration. Our approach encodes volumetric blocks to capture and differentiate underlying structures. It further incorporates a CLIP Score mechanism, which provides semantic information to the blocks to guide navigation. The navigation is empowered by a reinforcement learning framework that leverage these semantic cues to efficiently search for and identify desired viewpoints that align with the user's intent. The selected viewpoints are evaluated using CLIP Score to ensure that they best reflect the user queries. By automating viewpoint selection, our method improves the efficiency of volumetric data navigation and enhances the interpretability of complex scientific phenomena.

Abstract:
We present VolMoVis, a method for dynamic tomographic reconstruction that supports real-time volume generation and volumetric motion visualization from 2D projections. Visualizing the motion of 3D anatomical structures, such as organs and tumors, is critical for computer-aided interventions. However, conventional 4D volumetric reconstruction methods typically produce a limited set of volumes at discrete phases, suffering from low temporal resolution. Moreover, it often requires extensive segmentation of 3D structures or regions for visualizing volumetric data, making it challenging to segment and visualize dynamic volumes in real-time. To address these challenges, VolMoVis framework employs a continuous implicit neural representation that decomposes the dynamic volumetric data into a static reference volume and a continuous deformation field. This decomposition, along with an efficient deformation network, enables our framework to achieve real-time volume generation and volumetric visualization of continuous anatomical motions. We evaluate VolMoVis on both 4D digital phantoms and real patient datasets, demonstrating its effectiveness for accurate anatomical reconstruction and motion tracking. Furthermore, we highlight its capabilities in real-time simultaneous volume generation and tumor segmentation for visualizing dynamic volumes and 4D tumor tracking, showcasing its potential in image-guided radiation therapy.

Abstract:
Advancements in volume visualization (VolVis) focus on extracting insights from 3D volumetric data by generating visually compelling renderings that reveal complex internal structures. Existing VolVis approaches have explored non-photorealistic rendering techniques to enhance the clarity, expressiveness, and informativeness of visual communication. While effective, these methods often rely on complex predefined rules and are limited to transferring a single style, restricting their flexibility. To overcome these limitations, we advocate the representation of VolVis scenes using differentiable Gaussian primitives combined with pretrained large models to enable arbitrary style transfer and real-time rendering. However, conventional 3D Gaussian primitives tightly couple geometry and appearance, leading to suboptimal stylization results. To address this, we introduce TexGS-VolVis, a textured Gaussian splatting framework for VolVis. TexGS-VolVis employs 2D Gaussian primitives, extending each Gaussian with additional texture and shading attributes, resulting in higher-quality, geometry-consistent stylization and enhanced lighting control during inference. Despite these improvements, achieving flexible and controllable scene editing remains challenging. To further enhance stylization, we develop image-and text-driven non-photorealistic scene editing tailored for TexGS-VolVis and 2D-lift-3D segmentation to enable partial editing with fine-grained control. We evaluate TexGS-VolVis both qualitatively and quantitatively across various volume rendering scenes, demonstrating its superiority over existing methods in terms of efficiency, visual quality, and editing flexibility.

Abstract:
In-depth analysis of competitive debates is essential for participants to develop argumentative skills and refine strategies, and further improve their debating performance. However, manual analysis of unstructured and unlabeled textual records of debating is time-consuming and ineffective, as it is challenging to reconstruct contextual semantics and track logical connections from raw data. To address this, we propose Conch, an interactive visualization system that systematically analyzes both what is debated and how it is debated. In particular, we propose a novel parallel spiral visualization that compactly traces the multidimensional evolution of clash points and participant interactions throughout debate process. In addition, we leverage large language models with well-designed prompts to automatically identify critical debate elements such as clash points, disagreements, viewpoints, and strategies, enabling participants to understand the debate context comprehensively. Finally, through two case studies on real-world debates and a carefully-designed user study, we demonstrate Conch's effectiveness and usability for competitive debate analysis.

Abstract:
Ownership relationships of early printed books from the 15th century reveal complex patterns of distribution and possession, offering valuable insights for historical research. This paper presents OwnershipTracker, a visual analytics application developed to explore and trace these relationships using data from the Material Evidence in Incunabula (MEI) database. OwnershipTracker integrates bibliographic records, copy-specific data, and book provenance and ownership details, enabling users to uncover intricate ownership sequences over time. The application combines several visualization techniques, including network graphs to map connections between owners, timelines for temporal analysis, chord diagrams to quantify transfer patterns, and a distinctive, collaboratively designed spiderweb-like diagram highlighting converging and dispersing ownership transfers through specific owners. Developed iteratively with input from historical book researchers, the application underwent multiple refinements to align with domain research requirements. A summative evaluation with domain experts showcased the tool's ability to address the defined requirements and tasks. The final version of OwnershipTracker is deployed and accessible at: https://booktracker.nms.kcl.ac.uk/ownership.

Abstract:
Embodiment shapes how users verbally express intent when interacting with data through speech interfaces in immersive analytics. Despite growing interest in Natural Language Interactions (NLIs) for visual analytics in immersive environments, users' speech patterns and their use of embodiment cues in speech remain underexplored. Understanding their interplay is crucial to bridging the gap between users' intent and an immersive analytic system. To address this, we report the results from 15 participants in a user study conducted using the Wizard of Oz method. We performed axial coding on 1,280 speech acts derived from 734 utterances, examining how analysis tasks are carried out with embodiment and linguistic features. Next, we measured Speech Input Uncertainty for each analysis task using the semantic entropy of utterances, estimating how uncertain users' speech inputs appear to an analytic system. Through these analyses, we identified five speech input patterns, showing that users dynamically blend embodied and non-embodied speech acts depending on data analysis tasks, phases, and Embodiment Reliance driven by the counts and types of embodiment cues in each utterance. We then examined how these patterns align with user reflections on factors that challenge speech interaction during the study. Finally, we propose design implications aligned with the five patterns.

Abstract:
This paper presents Visualization Badges, graphical labels shown alongside visualizations to communicate provenance and design considerations to enhance understandability and transparency. Badges may, for example, highlight a major finding, disclose that an axis has been truncated, or warn of possible visual artifacts. Inspired by nutrition and energy labels on product packaging, visualization badges aim (i) to allow visualization authors to justify and disclose analysis and design decisions and (ii) to make readers aware of important information when viewing and interpreting visualizations. Collectively, visualization badges aim to foster trust in visualizations and prevent readers from drawing incorrect conclusions. Based on a series of co-design workshops, we define and evaluate the concept of visualization badges and formulate a conceptual framework for analysis, application, and further research. Our framework includes a catalog of 132 visualization badges, categorization schemes, design options for their visual representations, applied visualization examples, and guidelines for their use. We hope that visualization badges will help communicate data and collectively improve communication, visualization literacy, and the quality of visualization techniques. Our badges, workshops, and guidelines can be found online https://vis-badges.github.io.

Abstract:
Constructing cell developmental trajectories is a critical task in single-cell RNA sequencing (scRNA-seq) analysis, enabling the inference of potential cellular progression paths. However, current automated methods are limited to establishing cell developmental trajectories within individual samples, necessitating biologists to manually link cells across samples to construct complete cross-sample evolutionary trajectories that consider cellular spatial dynamics. This process demands substantial human effort due to the complex spatial correspondence between each pair of samples. To address this challenge, we first proposed a GNN-based model to predict cross-sample cell developmental trajectories. We then developed TrajLens, a visual analytics system that supports biologists in exploring and refining the cell developmental trajectories based on predicted links. Specifically, we designed the visualization that integrates features on cell distribution and developmental direction across multiple samples, providing an overview of the spatial evolutionary patterns of cell populations along trajectories. Additionally, we included contour maps superimposed on the original cell distribution data, enabling biologists to explore them intuitively. To demonstrate our system's performance, we conducted quantitative evaluations of our model with two case studies and expert interviews to validate its usefulness and effectiveness.

Abstract:
Aortic dissection is a life-threatening cardiovascular disease characterized by blood entering the media layer of the aortic vessel wall. This creates a second flow channel, known as the false lumen, which weakens the aortic wall and can potentially lead to fatal aortic rupture. Current risk stratification of aortic dissections is primarily based on morphological features of the aorta. However, hemodynamics also play a significant role in disease progression, though their investigation and visualization remain challenging. Common flow visualizations often experience visual clutter, especially when dealing with the intricate morphologies of aortic dissections. In this work, we implement and evaluate different approaches to visualizing the flow in aortic dissections effectively. We employ three techniques, namely streaklines with depth-dependent halos, transparent streaklines, and smoke surfaces. The latter is a technique based on streak surfaces, enhanced with opacity modulations, to produce a smoke-like appearance that improves visual clarity. We adapt the original opacity modulation of smoke surfaces to visualize flow even within the complex geometries of aortic dissections, thereby enhancing visual fidelity. To effectively capture dissection hemodynamics, we developed customized seeding structures that adapt to the shape of the surrounding lumen. Our evaluation, conducted via an online questionnaire, included medical professionals, fluid simulation experts, and visualization specialists. By analyzing results across these groups, we highlight differences in preference and interpretability, offering insight into domain-specific needs. No single visualization technique emerged as the best overall. Smoke surfaces provide the best overall clarity and visual realism. However, participants found streaklines with halos to be the best for quantifying flow, dispite them introducing significant visual clutter. Transparent streaklines serve as a middle ground, offering improved clarity over halos while maintaining some level of detail. Across all participant groups, smoke surfaces were rated as the most visually appealing and lifelike, with medical professionals highlighting their resemblance to contrast-agent injections used in clinical practice.

Abstract:
Understanding how individuals interpret charts is a crucial concern for visual data communication. This imperative has motivated a number of studies, including past work demonstrating that causal priors—a priori belief about causal relationships between concepts—can have significant influences on the perceived strength of variable relationships inferred from visualizations. This paper builds on these previous results, demonstrating that causal priors can also influence the types of patterns that people perceive as the most salient within ambiguous scatterplots that have roughly equal evidence for trend and cluster patterns. Using a mixed-design approach that combines a large-scale online experiment for breadth of findings with an in-person think-aloud study for analytical depth, we investigated how users’ interpretations are influenced by the interplay between causal priors and the visualized data patterns. Our analysis suggests two archetypal reasoning behaviors through which people often make their observations: contextualization, in which users accept a visual pattern that aligns with causal priors and use their existing knowledge to enrich interpretation, and rationalization, in which users encounter a pattern that conflicts with causal priors and attempt to explain away the discrepancy by invoking external factors, such as positing confounding variables or data selection bias. These findings provide initial evidence highlighting the critical role of causal priors in shaping high-level visualization comprehension, and introduce a vocabulary for describing how users reason about data that either confirms or challenges prior beliefs of causality.

Abstract:
Creating detailed 3D human avatars with fitted garments traditionally requires specialized expertise and labor-intensive workflows. While recent advances in generative AI have enabled text-to-3D human and clothing synthesis, existing methods fall short in offering accessible, integrated pipelines for generating CG-ready 3D avatars with physically compatible outfits; here we use the term CG-ready for models following a technical aesthetic common in computer graphics (CG) and adopt standard CG polygonal meshes and strands representations (rather than radiance field representations like NeRF and 3DGS) that can be directly integrated into conventional CG pipelines and support downstream tasks such as physical simulation. To bridge this gap, we introduce Tailor, an integrated text-to-3D framework that generates high-fidelity, customizable 3D avatars dressed in simulation-ready garments. Tailor consists of three stages. (1) Semantic Parsing: we employ a large language model to interpret textual descriptions and translate them into parameterized human avatars and semantically matched garment templates. (2) Geometry-Aware Garment Generation: we propose topology-preserving deformation with novel geometric losses to generate body-aligned garments under text control. (3) Consistent Texture Synthesis: we propose a novel multi-view diffusion process optimized for garment texturing, which enforces view consistency, preserves photorealistic details, and optionally supports symmetric texture generation common in garments. Through comprehensive quantitative and qualitative evaluations, we demonstrate that Tailor outperforms state-of-the-art methods in fidelity, usability, and diversity.

Affiliations: School of Computer Science and Technology, Zhejiang Gongshang University, Hangzhou, China; School of Artificial Intelligence and Computer Science, Nantong University, Nantong, China; Department of Computer Science, College of Computer Engineering and Sciences, Prince Sattam Bin Abdulaziz University, AlKharj, Saudi Arabia; Department of Computing Sciences, Tampere University, Tampere, Finland; Department of Computer Science, Universidade Federal Fluminense, Niteroi, Brazil; Department of Electrical and Electronic Engineering, Faculty of Technology, Firat University, Elazig, Türkiye; Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland

Abstract:
The automatic design of a 3D tooth model plays a crucial role in dental digitization. However, current approaches face challenges in compositional 3D tooth generation because both the layouts and shapes of missing teeth need to be optimized. In addition, collision conflicts are often omitted in 3D Gaussian–based compositional 3D generation, where objects may intersect with each other due to the absence of explicit geometric information on the object surfaces. Motivated by graph generation through diffusion models and collision detection using 3D Gaussians, we propose an approach named DM-CFO for compositional tooth generation, where the layout of missing teeth is progressively restored during the denoising phase under both text and graph constraints. Then, the Gaussian parameters of each layout-guided tooth and the entire jaw are alternately updated using score distillation sampling (SDS). Furthermore, a regularization term based on the distances between the 3D Gaussians of neighboring teeth and the anchor tooth is introduced to penalize tooth intersections. Experimental results on three tooth-design datasets demonstrate that our approach significantly improves the multiview consistency and realism of the generated teeth compared with existing methods.

Abstract:
Problem solving is a composite cognitive process, invoking a number of cognitive mechanisms, such as perception and memory. Individuals may form collectives to solve a given problem together in collaboration, especially when complexity is perceived to be high. To determine if and when collaborative problem solving is desired in the context of visual graph analysis, we compare ad hoc pairs to individuals and nominal pairs, when solving different tasks in mixed reality. We discuss the results of an experiment with 72 participants performed in two countries and three languages. We apply the concept of task instance complexity to quantify the visual demand of tasks used in the experiment. Our results show the importance of using nominal groups as a benchmark for evaluating collaborative virtual environments. We conclude that 3D graph representation is not sufficient to induce better collaborative results compared to the benchmark.

Abstract:
Volumetric data arises in many scientific disciplines, describing the tissue in our body, the composition of the Earth, or the distribution of matter in the universe, to name a few. Such volume data is commonly visualized with direct methods such as ray marching, and with indirect methods such as isocontours. The combination of these approaches is fruitful to provide context and to enhance depth perception. A common challenge, however, is that three-dimensional data inherently leads to occlusions, for which visibility optimization techniques have been employed in the past. The joint visibility optimization of volumetric data (direct volume rendering) and surface geometry (indirect volume rendering) is challenging. With this paper, we provide an optimization approach in which explicit and implicit geometry, such as isocontours or closed context geometry, as well as the volume itself give way to reveal structures that have been identified as important by a user. We model the geometry implicitly as level set of a signed distance field, which is evolved under a normal flow to reduce the occlusion. The non-linear optimization involves gradient descent solvers, level set propagation, and multi-grid optimization. We compare our approach to previous visibility optimizations.

Abstract:
We present DIQ-MPM, a novel monolithic two-way coupling framework for simulating interactions between solids modeled with the total Lagrangian formulation and Eulerian incompressible fluids using the Material Point Method (MPM). Our approach combines an implicit TLMPM formulation with a mixed velocity-pressure scheme to robustly simulate compressible solids undergoing large deformations, while eliminating numerical fractures. To enable strong fluid–solid coupling without relying on overlapping grids, we introduce a Dual Interface Quadrature (DIQ) mechanism that maps fluid-solid interface information consistently between the current and reference configurations. This allows us to construct a unified sparse pressure-only system via Schur complement, leading to efficient and stable coupling. We also integrate a particle-based contact force model to resolve solid-solid and solid-boundary contacts within implicit TLMPM. Experimental results demonstrate that our method stably captures free-slip coupling, large deformation phenomena, and complex interactions between compressible solids and incompressible fluids.

Abstract:
Generating realistic two-person interaction motions from text holds immense potential in computer vision and animations. While existing latent motion diffusion models offer compact and efficient representations, they often fail to produce physically plausible contacts and are typically constrained to a single canonical body shape. As a result, the generated motion sequences exhibit substantial mesh penetrations and lack interaction realism. To address these limitations, we propose a contact and shape-aware latent motion representation and diffusion model (CoShMDM) for generating realistic two-person interactions from text. Our framework begins by constructing contact-compatible motion using SMPL-based meshes and a normal alignment-based mesh contact matrix to capture fine-grained mesh-level contacts. To account for shape diversity, we incorporate SMPL shape parameters and iteratively learn contact dynamics across different body shapes. Additionally, a reinforcement learning-based mesh penetration avoidance policy network, guided by signed distance fields, is introduced to minimize mesh penetrations while preserving contact fidelity and shape-aware motion. We further employ a dual-encoder VQ-VAE to learn disentangled latent representations for motion and contacts, which are then utilized in a text- and body-shape-conditioned diffusion model. To ensure spatial, temporal, and semantic coherence, we integrate a novel contact and motion consistency module into the diffusion transformer. Extensive evaluations on the InterHuman and InterX datasets demonstrate that our method outperforms state-of-the-art approaches achieving lowest FID scores (4.801 and 0.013), with 19% and 17.3% reductions in mesh penetrations, and 17.8% and 33.2% gains in contact similarity, respectively.

Abstract:
Photo-realistic scene reconstruction from sparse-view, uncalibrated images is highly required in practice. Although some successes have been made, existing methods are either Sparse-View but require accurate camera parameters (i.e., intrinsic and extrinsic), or SfM-free but need densely captured images. This paper proposes Dust to Tower (D2T), a novel coarse-to-fine framework to address the coupled difficulty. The key idea is to explicitly narrow down the solution space and then introduce reliable supervision at novel viewpoints without resorting to expensive diffusion-based view synthesis. To do this, we first introduce a Coarse Construction Module (CCM) which exploits a fast Multi-View Stereo model to initialize a 3D Gaussian Splatting (3DGS) and recover initial camera poses. To refine the 3D model at novel viewpoints, we introduce Confidence-Aware Depth Alignment (CADA), which aligns a monocular inverse-depth prior to the reliable regions of the coarse depth using DUSt3R confidence, producing sharp and scale-consistent depth maps for accurate warping. We further propose Warped Image-Guided Inpainting (WIGI), which converts the accurate warped views into multi-view-consistent pseudo supervision via elaborate warping and inpainting process. Experiments on three benchmark datasets show that D2T achieves superior novel view synthesis quality and pose accuracy over ten representative baselines, while keeping high efficiency.

Abstract:
Our work contributes to the fast-growing literature on the use of Large Language Models (LLMs) to perform graph-related tasks. In particular, we focus on usage scenarios that rely on the visual modality, feeding the model with a drawing of the graph under analysis. We investigate how the model's performance is affected by the chosen layout paradigm, the aesthetics of the drawing, and the prompting technique used for the queries. We formulate three corresponding research questions and present the results of a thorough experimental analysis. Our findings reveal that choosing the right layout paradigm and optimizing the readability of the input drawing from a human perspective can significantly improve the performance of the model on the given task. Moreover, selecting the most effective prompting technique is a challenging yet crucial task for achieving optimal performance.

Abstract:
Rendering complex dynamic scenes offers the advantage of observing and understanding the real world. However, existing Dynamic Scene Rendering (DSR) methods remain challenged by suboptimal reconstruction fidelity. These limitations stem from relying on a single, unified deformation model, which struggles to capture complex motions involving multiple sub-motions and abrupt geometric transitions. While temporal decomposition methods could alleviate such shortcomings, they introduce the additional challenge of ignoring motion correlations and increasing storage requirements. To address these issues, we introduce Keyframe-driven Predictable Gaussian Splatting (KPGS)—an efficient framework for high-fidelity complex dynamic scene rendering. First, we present a patch-wise HSV clustering for extracting keyframes. Second, a prediction network based on the Transformer is utilized to calculate the deformable Gaussians at discrete keyframe times via voxelization. Third, we propose an inter-frame deformation network and a mutual supervision between adjacent segments to maintain the temporal continuity. Extensive experiments on our newly built dataset (MotionGS), as well as public benchmarks HyperNeRF and Neu3D, demonstrate that KPGS could achieve a higher average view synthesis performance than SOTA approaches, while maintaining a balance between storage cost and performance. More details of the demo and dataset are available at KPGS Supplementary.

Abstract:
The labor- and experience-intensive creation of 3D assets with physically based rendering (PBR) materials demands an autonomous 3D asset creation pipeline. However, most existing 3D generation methods focus on geometry modeling, either baking textures into simple vertex colors or leaving texture synthesis to post-processing with image diffusion models. To achieve end-to-end PBR-ready 3D asset generation, we present Lightweight Gaussian Asset Adapter (LGAA), a novel framework that unifies the modeling of geometry and PBR materials by exploiting multi-view (MV) diffusion priors from a novel perspective. The LGAA features a modular design with three components. Specifically, the LGAA Wrapper reuses and adapts network layers from MV diffusion models, which encapsulate knowledge acquired from billions of images, enabling better convergence in a data-efficient manner. To incorporate multiple diffusion priors for geometry and PBR synthesis, the LGAA Switcher aligns multiple LGAA Wrapper layers encapsulating different knowledge. Then, a tamed variational autoencoder (VAE), termed LGAA Decoder, is designed to predict 2D Gaussian Splatting (2DGS) with PBR channels. Finally, we introduce a dedicated post-processing procedure to effectively extract high-quality, relightable mesh assets from the resulting 2DGS. Extensive quantitative and qualitative experiments demonstrate the superior performance of LGAA with both text- and image-conditioned MV diffusion models. Additionally, the modular design enables flexible incorporation of multiple diffusion priors, and the knowledge-preserving scheme effectively preseves the 2D priors learned on massive image dataset, which leads to data efficient finetuning to lift the MV diffuison models for 3D generation with merely 69 k multi-view instances.

Abstract:
Fast simulation of elastic bodies is fundamental to computer graphics, yet current leading methods have a key limitation: they require fixed mesh connectivity. Existing methods leverage this assumption to achieve high performance but fail during topological changes such as cutting, fracturing, or merging. We present CAN, a novel optimizer that fundamentally decouples simulation acceleration from mesh topology. CAN introduces two Hessian-free, curvature-aware components: a Curvature-Aware Momentum (CAM) scheme that prevents overshooting by adaptively decaying momentum based on local gradient variations, and a Curvature-Aware Line Search (CALS) that provides high-quality step sizes via efficient directional curvature approximations. Since CAN relies solely on per-vertex, historical information, it is inherently parallel and topology-agnostic. We demonstrate that CAN achieves superior convergence compared to prior works across a wide range of dynamic-topology scenarios without any precomputation tied to connectivity, establishing a new paradigm for robust and efficient physics-based animation.

Abstract:
To study complex real-world phenomena using computer simulations, scientists often rely on ensemble datasets generated from multiple simulation runs with varying parameter configurations. This process can produce ensemble datasets with many members, making traditional data analysis pipelines impractical due to limited I/O bandwidth and disk capacity. Distribution-based data representations have been proposed as a promising solution. Processing data in situ to generate compact distribution-based representations not only alleviates the challenges of limited I/O bandwidth and disk capacity but also enables uncertainty quantification, thus mitigating the risk of misinterpretation. Nevertheless, distribution-based methods inherently sacrifice spatial information of data samples within the distribution, potentially reducing precision in the data analysis pipeline. To address this issue, we introduce a deep learning model to reconstruct data volume from the distribution representation. Instead of using a model that predicts a data block directly from its distribution representation, we propose a deep learning model based on the Sinkhorn operator and Gumbel trick that learns to map samples drawn from a distribution to spatial locations within the block. The deep learning model can support high-quality downstream data analysis and visualization, provide point-wise uncertainty quantification, and guarantee the distribution of the reconstructed data block follows the block’s distribution representation.

Abstract:
The advent of Retrieval-Augmented Generation (RAG) has significantly enhanced the ability of Large Language Models (LLMs) to produce factually accurate and up-to-date responses. However, the performance of a RAG system is not determined by a single component but emerges from a complex interplay of modular choices, such as embedding models and retrieval algorithms. This creates a vast and often opaque configuration space, making it challenging for developers to understand performance trade-offs and identify optimal designs. To address this challenge, we present RAGExplorer, a visual analytics system for the systematic comparison and diagnosis of RAG configurations. RAGExplorer guides users through a seamless macro-to-micro analytical workflow. Initially, it empowers developers to survey the performance landscape across numerous configurations, allowing for a high-level understanding of which design choices are most effective. For a deeper analysis, the system enables users to drill down into individual failure cases, investigate how differences in retrieved information contribute to errors, and interactively test hypotheses by manipulating the provided context to observe the resulting impact on the generated answer. We demonstrate the effectiveness of RAGExplorer through detailed case studies and user studies, validating its ability to empower developers in navigating the complex RAG design space. Our code and user guide are publicly available at https://github.com/Thymezzz/RAGExplorer.

Abstract:
Large Language Models (LLMs) demonstrate exceptional capabilities in factual question answering, yet they sometimes provide incorrect responses. To address this issue, knowledge editing techniques have emerged as effective methods for correcting factual information in LLMs. However, typical knowledge editing workflows struggle with identifying the optimal set of model layers for editing and rely on summary indicators that provide insufficient guidance. This lack of transparency hinders effective comparison and identification of optimal editing strategies. In this paper, we present KEditVis, a novel visual analytics system designed to assist users in gaining a deeper understanding of knowledge editing through interactive visualizations, improving editing outcomes, and discovering valuable insights for the future development of knowledge editing algorithms. With KEditVis, users can select appropriate layers as the editing target, explore the reasons behind ineffective edits, and perform more targeted and effective edits. Our evaluation, including usage scenarios, expert interviews, and a user study, validates the effectiveness and usability of the system.

Abstract:
Root cause analysis (RCA) is critical for maintaining the availability and efficiency of cloud computing systems. However, identifying root causes from the large-scale, high-dimensional monitoring data generated by these complex environments is a significant challenge. Current approaches often rely on time-consuming manual analysis to ensure flexibility and reliability, while recent automated methods lack the crucial insights provided by domain experts. To bridge this gap, we propose RCInvestigator, a visual analytics system that facilitates interactive root cause investigation by establishing a tight collaboration between human experts and machine analysis. Our approach addresses three key challenges: a) modeling databases for the root cause investigation, b) inferring root causes from large-scale time series, and c) building comprehensible investigation results. We demonstrate the effectiveness and utility of RCInvestigator through two real-world case studies, which received positive feedback from domain experts.

Abstract:
Scenes are continuously undergoing dynamic changes in the real world. However, existing human-scene interaction generation methods typically treat the scene as static, which deviates from reality. Inspired by world models, we introduce Dyn-HSI, the first cognitive architecture for dynamic human-scene interaction, which endows virtual humans with three humanoid components. (1) Vision (human eyes): we equip the virtual human with a Dynamic Scene-Aware Navigation, which continuously perceives changes in the surrounding environment and adaptively predicts the next waypoint. (2) Memory (human brain): we equip the virtual human with a Hierarchical Experience Memory, which stores and updates experiential data accumulated during training. This allows the model to leverage prior knowledge during inference for context-aware motion priming, thereby enhancing both motion quality and generalization. (3) Control (human body): we equip the virtual human with Human-Scene Interaction Diffusion Model, which generates high-fidelity interaction motions conditioned on multimodal inputs. To evaluate performance in dynamic scenes, we extend the existing static human-scene interaction datasets to construct a dynamic benchmark, Dyn-Scenes. We conduct extensive qualitative and quantitative experiments to validate Dyn-HSI, showing that our method consistently outperforms existing approaches and generates high-quality human-scene interaction motions in both static and dynamic settings.

Abstract:
Virtual locomotion remains a challenge in VR, especially in space-limited environments where room-scale walking is impractical. We present LocoScooter, a low-cost, deployable locomotion interface combining foot-sliding on a compact treadmill with handlebar steering inspired by scooter riding. Built from commodity hardware, it supports embodied navigation through familiar, physically engaging movement. In a within-subject study (N = 14), LocoScooter significantly improved immersion, enjoyment, and bodily involvement over joystick navigation, while maintaining comparable efficiency and usability. Despite higher physical demand, users did not report increased fatigue, suggesting familiar movements can enrich VR navigation.

Abstract:
We introduce SceneLinker, a novel framework that generates compositional 3D scenes via semantic scene graph from RGB sequences. To adaptively experience Mixed Reality (MR) content based on each user's space, it is essential to generate a 3D scene that reflects the real-world layout by compactly capturing the semantic cues of the surroundings. Prior works struggled to fully capture the contextual relationship between objects or mainly focused on synthesizing diverse shapes, making it challenging to generate 3D scenes aligned with object arrangements. We address these challenges by designing a graph network with cross-check feature attention for scene graph prediction and constructing a graph-variational autoencoder (graph-VAE), which consists of a joint shape and layout block for 3D scene generation. Experiments on the 3RScan/3DSSG and SG-FRONT datasets demonstrate that our approach outperforms state-of-the-art methods in both quantitative and qualitative evaluations, even in complex indoor environments and under challenging scene graph constraints. Our work enables users to generate consistent 3D spaces from their physical environments via scene graphs, allowing them to create spatial MR content. Project page is https://scenelinker2026.github.io.

Abstract:
Conventional Augmented Reality (AR) head-mounted displays (HMDs) integrate all essential components within the wearable unit, imposing limitations on weight, heat dissipation, and battery life. Beaming Displays (BDs) address these constraints by externalizing computation and projection; however, achieving reliable low-latency six-degrees-of-freedom (6-DoF) tracking remains a challenge. We present BeamStellar, a BD system that enables low-latency 6-DoF head tracking by combining a position-sensitive detector (PSD) with a spatio-temporally modulated infrared LED array mounted on passive glasses. The system integrates tracking and projection via a co-aligned optical path and performs real-time signal processing on an FPGA to minimize tracking latency. We describe the system architecture, pose reconstruction pipeline, and evaluate the pose-tracking latency of approximately 430 µs. BeamStellar provides a replicable, lightweight AR tracking platform, establishing a new paradigm for low-latency environment-based AR systems.

Abstract:
Prior research has developed a number of manipulation techniques that can achieve precise object placement in virtual reality, but studies of these techniques typically use simple objects. We conducted a study comparing two existing techniques, (AMP-IT and WISDOM), during alignment of objects with complex geometry to evaluate the potential influence of geometric complexity on performance, usability, workload and preference. Our findings indicate that participants had faster completion times and higher trial completion rates with AMP-IT on high-precision alignment tasks, contrary to earlier findings that used simple objects. Yet WISDOM is still preferred and considered more usable, despite increased workload and poorer performance, exposing participants' willingness to trade objective performance for comfort during use.

Abstract:
The poor transparency of occlusion-capable optical see-through head-mounted displays (OC-OSTHMDs) deteriorates the visibility of the real scene, hindering the practical application of the devices. Previous works mitigate the issue by upgrading the transmittance of the spatial light modulator (SLM). However, the strategy soon reaches a limit because further optimization requires improving the transmittance of all optical elements, e.g., lenses and beam splitters. Moreover, pixelated occlusion usually relies on polarizing the real scene light, inevitably cutting the input optical power by half. To overcome this limitation, we propose a mask balancing method that improves real-scene brightness through polarization blending. Specifically, the s-polarized component, which passes through the optical system to provide occlusion-capable vision, is blended with the p-polarized component, which bypasses the system to preserve the raw view of the real scene. The blending is realized by simply modulating the cross-angle between a polarizing beam splitter and a linear polarizer, benefiting the robustness and versatility of the proposed method. We introduce a perception-driven blending approach, where the cross-angle is optimized in real-time to balance the visibility of the real scene and the texture and lighting of the virtual object. A benchtop prototype is built. A user study with 12 participants is conducted to quantify the visibility threshold of the texture and lighting of virtual objects. Then, a user study with 12 participants proves that the proposed method improves the visibility of the real scene while keeping a good appearance of the virtual object. We believe the proposed method is an important step toward developing practical solutions for OC-OSTHMDs.

Abstract:
Image-guided percutaneous needle puncture is a critical yet challenging clinical procedure, constrained by the high cognitive demand of mental model construction and manipulation of 3D anatomy via scrolling through 2D cross-sectional images. While virtual reality (VR) simulators provide a risk-free training platform, many focus on simulation fidelity but lack structured, self-directed learning frameworks. In this paper, we present PunctVR, a VR system that incorporates the instructional principles of scaffolding. PunctVR features a training mode employing a phased subgoal workflow and instructional guidance scaffolding, and an assessment-only test mode where both the workflow and enhanced 3D visualization are removed. We conducted a between-subject experiment with 16 physicians, comparing training using a baseline multiplanar reconstruction (MPR) view with a combined MPR + 3D visualization across two difficulty levels. Our test mode results indicate that all trainees significantly improved their performance after training. Furthermore, those who trained with the integrated 3D visualization achieved a greater reduction in puncture time in both easy and hard cases. These findings suggest that PunctVR effectively enhances procedural efficiency in simulated needle puncture training and provides important insights into how learning scaffolding can accelerate skill acquisition and retention for image-guided interventions.

Abstract:
Motion tracking to project users into embodied virtual reality (VR) as avatars is an essential application of real-time computer graphics. Most current embodied VR systems rely on head-mounted displays (HMDs) to estimate user pose, as headset sensors can track the head and hands, thereby reconstructing the full body without the need for external hardware. However, measuring the quality of motion reconstruction algorithms from HMD-based tracking, particularly those intended for use in social settings, remains challenging due to the complex interaction between motion and perceived social signals. This paper compares two industrial tracking reconstruction solutions, called HMD1 (i.e., a basic HMD-based method that uses head tracking and hand positions estimated from HMD cameras) and HMD2 (i.e., an advanced HMD-based method with additional onboard camera streams), that estimate user motion using only an HMD against ground-truth motion capture (MoCap) data. It advocates for a social signal-based analysis that views motion as a communication medium and employs user observations to measure whether viewers successfully perceive the information encoded in motion. Across 156 socially expressive clips, Social Signal ratings were more effective than generic measures at revealing differences between the HMD methods. HMD2 preserved social meaning more accurately than HMD1, with fewer significant deviations from MoCap, while both HMD methods were frequently rated less natural than MoCap. A qualitative review localized recurrent failure modes, such as arm swivel/shoulder errors, posture reconstruction issues, and floating/stance artifacts, which help explain the misreading of social signals. We release a dashboard scorecard, motion capture data, and a benchmark protocol to enable consistent motion evaluation. More generally, this work advocates for an underexplored approach to motion evaluation that focuses on assessing the semantics of motion to determine quality. As reliance on generative artificial intelligence (AI) increases, it is essential to standardize evaluation to preserve the authenticity of the social signals conveyed. The developed dataset and the evaluation framework are provided on our project's website: https://github.com/facebookresearch/MotionIsTheMessageDataset.

Abstract:
Extended Reality (XR) technologies offer many possibilities to support collaboration between remote and local team members. However, in a scenario with multiple devices and multiple interaction options, including virtual, augmented and the continuum of realities among them, it is still unclear how to select the right option to make collaboration more efficient. To provide some design insights in this context, this paper presents a comparative study of collaboration behaviours and outcomes when using different XR technologies, where X stands for Virtual or Augmented. In our experimental setup, pairs of users collaborated on three tasks drawn from visual analytics, each designed to require a different level of synchronization: low, medium, and high. We conducted a study with 40 participants (20 dyads), who performed tasks in two conditions: Augmented Reality (AR) and Virtual Reality (VR). All used the same equipment and interactions; only the immersive environment differed. The results suggest that collaboration outcomes differed only in high-coupling tasks, with AR showing a performance advantage. This may stem from a greater difficulty of synchronizing actions when perceiving a partner through avatars rather than directly. However, no significant differences were observed in collaboration dynamics, and the choice of XR technology did not appear to affect collaboration patterns. These findings suggest that, for collaborative analytical tasks comparable to those in this study, collaborators can choose the XR mode that best fits their preferences and current needs without concern about hindering teamwork.

Abstract:
Mid-air gestures in Extended Reality (XR) often lead to fatigue, discomfort and imprecision, limiting their suitability for extended use. Surface-based interactions offer a compelling alternative, providing improved accuracy, speed, and comfort. However, current egocentric vision-based methods struggle with reliable surface inputs due to challenges in hand tracking and surface plane estimation from oblique and occluded viewing angles. To this extent, we introduce SurfaceXR, a novel sensor fusion approach that combines headset based hand tracking with micro-vibration data sampled from commodity smartwatch IMUs to enable precise and robust inputs on everyday surfaces. Our system is designed with flexibility in mind — it can function using only hand tracking, only IMU sensing, or optimally with both modalities combined, and remains robust even without explicit surface calibration. Our key insight is that these modalities are complementary — hand tracking provides 3D positional data of hand joints, whereas IMUs supply high-frequency wrist/hand motion data. Our user study across 21 participants validates SurfaceXR's effectiveness in augmenting surface touch tracking and 8 class hand-surface gesture recognition, demonstrating significant improvements over single-modality approaches. Enabled by SurfaceXR, we demonstrate a series of interactive apps for both AR and VR, ranging from on-surface sketching, text entry and gesture-based navigation.

Abstract:
In this paper, we present FanType, a text entry technique that enables efficient typing on small virtual keyboards in extended reality (XR), closely resembling thumb-based typing on smartphones and tablets. While virtual environments theoretically provide unlimited space, practical XR usage often occurs in mobile and spatially constrained contexts, such as when using augmented reality (AR) headsets while walking or in confined environments, where large virtual keyboards are impractical, as they occupy substantial visual space and interfere with other virtual or real content. To address this, FanType integrates a fan-shaped interface that groups keys within the thumb's reach and leverages a lift-up gesture for disambiguation, thereby reducing hand movement while preserving natural typing speed. Additionally, an intention-inferring mechanism involving tailored algorithms is implemented to counterbalance touch bias and further ensure input accuracy. A user study with 22 participants evaluated the method on tablet- and smartphone-size portrait keyboards, comparing it with a baseline (touch-only) and a state-of-the-art mid-air technique. Results provide insights into the performance of such a design and highlight the challenges of designing typing methods for small virtual keyboards.

Abstract:
We propose a spatial affordance-aware affine transformation method between heterogeneous spaces for continuous multi-object matching in shared Mixed Reality (MR) spaces. While previous redirection and spatial mapping approaches utilize physical objects and walkable areas, a critical gap remains in enabling continuous mapping between dissimilar physical environments that supports both precise object alignment and seamless locomotion in a shared space. Our method structurally segments heterogeneous spaces into interaction zones and constructs affine patches based on object adjacency and facing configuration, enabling continuous correspondence. We evaluate our method using a dataset of paired dissimilar spaces and demonstrate that, unlike conventional grid-based methods, our approach achieves broader spatial alignment and richer object matching. The results show that our method can serve as an effective mapping framework for shared environments requiring semantic continuity and structural coherence across diverse real-world spaces.

Abstract:
Controlling multiple agents in parallel is an important challenge in human-robot interaction, with growing applications in remote collaboration, telepresence, and multi-agent coordination. However, empirical work on distributed embodiment remains limited, particularly in scenarios where independently controlled avatars operate in distinct environments. To address this gap, we propose the concept of Dual Body Sensation (DBS): a novel theoretical construct of distributed embodiment, describing a state in which a single operator simultaneously experiences embodiment for two independent robots in distinct environments during their control. We conducted a within-subject experiment to investigate the cognitive feasibility of DBS using a 2× 3 factorial design that varied visual presentation (independent view, superimposed view) and tactile presentation (no tactile, bilateral tactile, unified tactile). The results indicated that DBS emerged across all conditions and was significantly enhanced under the unified tactile condition. The combination of the superimposed view and unified tactile conditions also increased the sense of body ownership and the sense of agency, while the superimposed view condition reduced cybersickness and cognitive load. These findings provide empirical support for DBS and inform the design of future multi-agent interaction systems that support flexible distribution of bodily control and awareness.

Abstract:
Fast, accurate spatial-temporal cognition is vital for situational awareness and mission success in indoor emergencies. Complex layouts and occlusions hinder first responders’ ability to judge distances, scales, and task completion times. This paper explores the enhancement of these cognitive abilities through the application of Augmented Reality (AR) devices. Utilizing mixed reality and reality capturing technologies, a digital twin model of a three-level indoor structure was built based on LiDAR scans of Turlington Hall at the University of Florida. A mixed reality environment was then constructed using the Unity engine and Microsoft HoloLens2. In a user study (N = 30), we systematically compared three visualization conditions (2D Map, 3D Map, and Egocentric Map) across spatial and temporal estimation tasks. Non-parametric statistical analysis with correction for multiple comparisons revealed a task-dependent pattern. The Egocentric Map showed the lowest error for ETA estimation (MAPE = 17.2% vs. 23.3% for 2D maps), suggesting a trend toward improved temporal cognition, though this pairwise difference did not reach statistical significance after correction. Conversely, for distance-between-objects tasks, 2D and 3D maps yielded significantly lower error than Egocentric views (MAPE = 24.3% and 17.7% vs. 36.8%, p_adj < 0.01padj<0.01). These findings reveal a task-dependent trade-off, indicating that no single visualization mode is universally optimal: AR overlays may support action-oriented temporal planning, while exocentric views remain superior for precise metric spatial judgments in emergency response.

Abstract:
3D Gaussian Splatting (3DGS) has been explored for surface reconstruction, however unstructured and discontinuous Gaussian point clouds lead to uneven surface reconstruction accuracy as well as frequent loss of Novel View Synthesis (NVS) quality. To address this problem, we propose a scale-aligned monocular depth-guided 3DGS, a promising novel framework that combines geometric prior regularization and consistency supervision to achieve high-quality rendering and surface reconstruction. Specifically, monocular depth, estimated by some recently proposed monocular depth estimation models, contain implicitly abundant valuable geometric cues, but scale ambiguity limits its application. Therefore we first propose a KK-Nearest Neighbor (KNN)-based depth alignment framework that utilizes the full-domain gradient at monocular depth map to align to the sparse point cloud obtained during the Structure from Motion (SfM), which is employed for regularization to enhance geometric representation. Then a pseudo-mesh-based multi-view consistency module is introduced to fine-tune and guide the model to recover the accurate surface. Finally, a pixel-level isotropic gradient aware method guides the appropriate growth of the Gaussians to further improve the surface and rendering quality. Experiments on dozens of indoor, outdoor, and object-centered/non-object-centered datasets demonstrate that our method achieves accurate surface reconstruction with excellent NVS performance.

Abstract:
With the acceleration of global ageing and the widespread adoption of electronic devices, modern visualization design must address the varying perceptual abilities across different age groups. Spatial frequency (SF) is commonly used as a stimulus dimension to assess perceptual ability in vision science, but its potential role as a guiding variable in visualization design, particularly for addressing age-related differences across the lifespan, remains largely unexplored. Many vision science studies have examined age-related spatial frequency perception, but with varying results. This review systematically analyses differences in experimental design and the reasons behind selecting experimental methods, which may be key to these differences. This study developed search terms and systematically searched the ACM, IEEE, PubMed, APA PsycINFO, SAGE, ScienceDirect, and Springer Nature databases. After rigorous screening, 33 studies were included and evaluated, with their similarities and differences compared across five dimensions. Cluster analysis further categorizes the literature into five groups, confirming that discrepancies in experimental outcomes largely stem from variations in psychological methods and participant characteristics. By considering the underlying reasons for these choices, we propose guidelines for designing perceptual experiments in which spatial frequency and age are manipulated factors. These recommendations provide theoretical support and practical insights for visualization design research that considers age-related perception of spatial frequency.

Abstract:
Existing quantitative evaluation metrics for point cloud denoising methods require both the denoised point cloud and the corresponding ground-truth clean point cloud to compute a representative geometric distance. This requirement is highly problematic in real-world scenarios, where ground-truth clean point clouds are often unavailable. In this paper, we propose a simple yet effective unsupervised geometric distance (UGD) for real-world noisy point cloud denoising, calculated solely from noisy point clouds. The core idea is to learn a patch-wise prior model from a set of clean point clouds and then employ this prior model as the ground-truth to quantify the degradation by measuring the geometric variations of the denoised point cloud. To this end, we first learn a pristine Gaussian Mixture Model (GMM) with extracted patch-wise quality-aware features from a set of pristine clean point clouds by a patch-wise feature extraction network, which serves as the ground-truth for the quantitative evaluation. Then, the UGD is defined as the weighted sum of distances between each patch of the denoised point cloud and the learned pristine GMM model in the patch space. To train the employed patch-wise feature extraction network, we propose a self-supervised training framework through multi-task learning, which includes pair-wise quality ranking, distortion classification, and distortion distribution prediction. Quantitative experiments with synthetic noise confirm that the proposed UGD achieves comparable performance to supervised full-reference metrics. Moreover, experimental results on real-world data demonstrate that the proposed UGD enables unsupervised evaluation of point cloud denoising methods based exclusively on noisy point clouds.

Abstract:
Eye-hand coordinated interaction is becoming a mainstream interaction modality in Virtual Reality (VR) user interfaces. Current paradigms for this multimodal interaction require users to learn predefined gestures and memorize multiple gesture-task associations, which can be summarized as an “Operation-to-Intent” paradigm. This paradigm increases users’ learning costs and has low interaction error tolerance. In this paper, we propose SIAgent, a novel “Intent-to-Operation” framework allowing users to express interaction intents through natural eye-hand motions based on common sense and habits. Our system features two main components: (1) intent recognition that translates spatial interaction data into natural language and infers user intent, and (2) agent-based execution that generates an agent to execute corresponding tasks. This eliminates the need for gesture memorization and accommodates individual motion preferences with high error tolerance. We conduct two user studies across over 60 interaction tasks, comparing our method with two “Operation-to-Intent” techniques. Results show our method achieves higher intent recognition accuracy than gaze + pinch interaction (97.2% versus 93.1%) while reducing arm fatigue and improving usability, and user preference. Another study verifies the function of eye gaze and hand motion channels in intent recognition. Our work offers valuable insights into enhancing VR interaction intelligence through intent-driven design.

Abstract:
We introduce a temporal arrangement framework for embedding visualizations in sports videos with a focus on swimming. Our work is inspired by strategies used in current TV broadcasts, where visualizations are selectively displayed to provide meaningful and engaging information to audiences. We began with a systematic review of TV broadcast practices, through which we identified recurring temporal combinations of visualizations and competition statuses, which we define as patterns of temporal arrangement for embedded visualizations. To move beyond the constraints of existing broadcast practices, we then conducted a formative study with a general population. Based on this broader perspective, we designed a configuration framework that allows us to formally specify when and for how long, related to swimming context metadata, visualizations appear in a video. We instantiate the framework in a technology probe, SwimChrono, for applications with real-world swimming context videos. Through audience-customized configurations, SwimChrono supports novel arrangements beyond those used in existing professional settings, is adaptable to various swimming contexts, including different lengths and swimming styles, and key events. Furthermore, we conduct user studies and contribute use cases to illustrate how our framework can be well applied for diverse needs.

Abstract:
Embedded data visualizations have emerged as a powerful narrative medium for conveying complex information within video footage. However, creating such content remains labor-intensive, as existing workflows rely on manual frame-by-frame adjustments to ensure spatial and temporal consistency. To address these challenges, we present ChartBlender, an interactive authoring system designed to streamline the creation, embedding, and automatic synchronization of data visualizations within video scenes. We develop a tracking pipeline that supports both object and camera tracking, ensuring robust alignment of visualizations with dynamic video content. To maintain visual clarity and aesthetic coherence, we also explore the design space of video-suited visualizations and develop a library of customizable templates optimized for video embedding. We evaluated ChartBlender through two controlled experiments and expert interviews with five domain experts. Results show that our system enables accurate synchronization and accelerates the production of data-driven videos.

Abstract:
There are many problem domains where users have to pay attention to a series of targets. In this article, we evaluated visual cues to support attention in dynamic situations in virtual environments. Three immersive visualizations (Emphasis, Deemphasis, Emphasis+Deemphasis) have been examined across four different attention types (sustained, alternating, divided, and selective attention) to support the user’s attention on targets. Emphasis highlights targets of interest using outlines; Deemphasis removes details with masks, and Emphasis+Deemphasis combines both techniques. An experiment with 34 participants examined the cues through four scenarios that required different attention types in dynamic situations. Results indicated that while Emphasis supported significantly faster reactions in most attention types, the other two visual cues showed superiority across other measurements. Although evaluated in a virtual environment, the findings offer guidance for designing visual cues in other dynamic situations that involve different attention types.

Abstract:
We propose Int3DNet, a scene-aware network that predicts 3D intention areas directly from scene geometry and head-hand motion cues, enabling robust human intention prediction without explicit object-level perception. In Mixed Reality (MR), intention prediction is critical as it enables the system to anticipate user actions and respond proactively, reducing interaction delays and ensuring seamless user experiences. Our method employs a cross attention fusion of sparse motion cues and scene point clouds, offering a novel approach that directly interprets the user's spatial intention within the scene. We evaluated Int3DNet on MoGaze and CIRCLE datasets, which are public datasets for full-body human-scene interactions, showing consistent performance across time horizons of up to 1500 ms and outperforming the baselines, even in diverse and unseen scenes. Moreover, we demonstrate the usability of proposed method through a demonstration of efficient visual question answering (VQA) based on intention areas. Int3DNet provides reliable 3D intention areas derived from head-hand motion and scene geometry, thus enabling seamless interaction between humans and MR systems through proactive processing of intention areas.

Abstract:
Endoscopic reconstruction and rendering technology are crucial for minimally invasive diagnosis and treatment. Due to physical constraints such as limited light source placement and narrow operational space, endoscopic scenes often contain some dark regions that compromise clinical observation and diagnostic accuracy. To address this issue, we propose AdaEndoGS for endoscopic 3D reconstruction with physically consistent dark region illumination enhancement. Specifically, built upon 3D Gaussian Splatting, AdaEndoGS enriches each Gaussian with surface attributes including normal, roughness, and reflectance, thereby constructing an illuminatable 3D scene representation. AdaEndoGS automatically identifies the dark region via ray tracing, leveraging Gaussian opacity, surface normal, and viewpoint information to adaptively plan the placement of a virtual light source. Supervised by a carefully designed loss function incorporating multiple illumination-related terms, AdaEndoGS optimizes light source intensity and attenuation parameters to generate harmonious illumination enhancement in the dark regions, while avoiding overexposure in originally bright areas. We comprehensively evaluate our method on the public dataset C3VD and a novel endoscopy simulation dataset created by us in Unity3D, which includes paired low-lit and well-lit images to facilitate quantitative evaluation. Experimental results show that AdaEndoGS more accurately simulates light-matter interactions compared to existing methods. It significantly improves visual quality and enhances detail visibility, offering an effective technical solution for advancing endoscopic image-based reconstruction and rendering. Project repository: https://webstermorton.github.io/adaendogs-website.

Abstract:
Engaging in rhythmic synchronized activities such as dancing or coordinated exercises is a powerful driver of social connection, as interpersonal entrainment promotes feelings of closeness, empathy, and cooperation. However, in media-based settings such as Social Virtual Reality (SVR), maintaining synchrony is technically challenging. Even modest latency can disrupt the temporal coordination needed for shared rhythm, thereby diminishing social presence, user engagement and the quality of interaction. Our work explores whether synchrony-reinforcing feedback can mitigate latency effects in a dyadic VR rowing task. Participants received real-time visual, audio, or audio-visual feedback: visual feedback was provided through particle effects, and audio feedback through music, both of which intensified with stronger synchronization and diminished as it weakened. Latency was systematically varied. Our findings show that latency, especially higher delays, significantly disrupts coordination and a sense of togetherness. Participants often misattribute these disruptions to personal or interpersonal factors. Audio-visual feedback reduces these effects, supporting smoother interaction and masking symptoms of latency. The task load remains stable across latency levels, but decreases when clear, multimodal feedback is provided. Notably, while closeness remains robust, togetherness deteriorates in the absence of timely, multimodal cues. Our results suggest that latency effects can be partially addressed through perceptual design. Providing multimodal feedback is a key strategy for maintaining synchrony, reducing cognitive effort, and sustaining meaningful social interaction in latency-prone environments.

Abstract:
Augmented Reality (AR) can enhance accessibility by anchoring virtual windows to the user's body. Among common approaches, head-following windows help maintain floating virtual windows within the user's field of view. Previous studies have actively explored this new design space to improve user experience and efficiency. In contrast, this study focuses on the perceived resistance of head-following windows in AR, despite their lack of physical mass. We conducted a within-subject experiment with 24 participants, manipulating Follow-Up Delay, Window Size, and UI Type. We measured subjective resistance ratings, NASA-TLX (Raw TLX Scores), and the gaze-head angular offset. The results showed that both a certain level of Follow-Up Delay and the Tag-Along elicited significantly stronger perceived resistance as well as task load. Although Window Size alone did not show a significant effect on resistance ratings, we observed an interaction between the size and UI Type. These findings extend existing pseudo-haptics research by revealing the previously unexplored domain of resistance in head-based interactions with head-following virtual windows. We further provide design implications for head-following windows in AR.

Abstract:
This study investigates Embodied Virtual Agents (EVAs) driven by Large Language Models (LLMs) as mediators in triadic collaboration where two users with conflicting goals must work towards a shared objective. We developed and evaluated an XR system where pairs (n=24) collaborated on an office design task, assessing the impact of agent presence and embodiment. Our findings reveal that agent embodiment significantly enhanced co-presence, which in turn fostered interactions that led to better perceived collaboration and higher user preference, whereas frequent interactions with the disembodied agent was associated with lower user preference. Conversely, agent presence did not improve task efficiency or satisfaction. Notably, our qualitative analysis revealed that the LLM-driven agent spontaneously adopted emergent facilitative and evaluative mediation strategies that align with collaborating and compromising conflict-resolving modes, showcasing its potential as an adaptive collaborative aid without explicit programming. These results highlight that the value of EVAs in complex collaboration lies in their ability to shape social dynamics and provide nuanced, context-aware mediation, a different form of value than traditional productivity enhancement.

Abstract:
Redirected walking (RDW) alters the relationship between physical locomotion and visual feedback to guide users along virtual paths that vary from their real-world trajectories. Over time these changes can lead to adaptation of the user and produce novel sensorimotor settings that users might retain across sessions and apply directly once they enter virtual reality (VR). Although short-term adaptation to altered sensorimotor contingencies is well established, it remains unclear whether such adaptation is retained across multiple days. We investigated long-term retention of RDW adaptation by repeatedly exposing ten participants to a fixed rightward curvature gain of π /30 across nine sessions over two weeks. Each session consisted of 200 walking repetitions with gain applied. Adaptation was assessed before and after each session using blind walking and a pointing task. Moreover, perceptual detection thresholds for curvature gains were measured before the first session (Baseline), a day after the ninth session (Final), and once in-between before the 5th adaptation session. Results showed clear retention of adapted locomotor behavior: during blind walking at the beginning of the sessions, when instructed to walk straight, participants consistently exhibited curved trajectories, which indicates that newly acquired sensorimotor contingencies can be retained over days and are immediately available upon re-entering VR. At the same time, pointing accuracy remained stable throughout the experiment and detection thresholds showed no consistent changes across sessions. In summary, our study provides evidence that adaptation to altered sensorimotor contingencies in VR can be retained across multiple sessions and days and can be available as soon as the user enters VR. This may be useful for many scenarios in which users repeatedly use VR tools over a long period of time. The complete data set, all supplemental materials and the preprint of the manuscript are available at https://osf.io/z973w.

Abstract:
We investigate how vibrotactile wrist feedback can enhance spatial guidance for handheld tool movement in optical see-through augmented reality (AR). While AR overlays are widely used to support surgical tasks, visual occlusion, lighting conditions, and interface ambiguity can compromise precision and confidence. To address these challenges, we designed a multimodal system combining AR visuals with a custom wrist-worn haptic device delivering directional and state-based cues. A formative study with experienced surgeons and residents identified key tool maneuvers and preferences for reference mappings, guiding our cue design. In a cue identification experiment (N = 21), participants accurately recognized five vibration patterns under visual load, with higher recognition for full-actuator states than spatial direction cues. In a guidance task (N = 27), participants using both AR and haptics achieved significantly higher spatial precision (5.8 mm) and usability (SUS = 88.1) than those using either modality alone, albeit with modest increases in task time. Participants reported that haptic cues provided reassuring confirmation and reduced cognitive effort during alignment. Our results highlight the promise of integrating wrist-based haptics into AR systems for high-precision, visually complex tasks such as surgical guidance. We discuss design implications for multimodal interfaces supporting confident, efficient tool manipulation.

Abstract:
Virtual Reality (VR) has been widely applied in healthcare domain. Virtual Reality Mirror Therapy (VRMT) is one such clinical application for phantom limb pain management and post-stroke hemiparesis rehabilitation. However, the prevailing design of VRMT tasks typically follows the schema of traditional mirror therapy paradigm that prescribes strictly symmetric bilateral movements along the body's midline. While symmetry is not a requirement by the mirror neuron therapy where mirror therapies derive, this constraint imposes unnecessary limitations to both the development of diversified game-based rehabilitation tasks and portability of VR equipment and hardware. As a result, average patients using a conventional VR system may often find it difficult to perform certain exercises, such as arm abductions, or not possible to run a VRMT paradigm on affordable VR equipment for home-based rehabilitation. We therefore propose AsymmVRMT, an asymmetric strategy for mirrored affected limb mapping, and adopted it in our VRMT paradigm. We conducted a randomized controlled trial (RCT) on post-stroke hemiparetic patients, and examined their clinical outcome measures (FM-UE, BI & WMFT) along with their user experience data (SUS & VEQ). Results showed that patients demonstrated stronger adaptability and more positive perceptions toward using the AsymmVRMT model, while the therapeutic outcomes were as good as those using conventional VRMT. AsymmVRMT therefore appears to be an effective alternative for human-computer interaction in VR that may offer higher flexibility and diversity to address the hardware-related constraints in rehabilitation system development, facilitating convenient VR-based rehabilitation for patients, both at the hospital and at home.

Abstract:
Trust plays a critical role in visual data communication and decision-making, yet existing visualization research employs varied trust measures, making it challenging to compare and synthesize findings across studies. In this work, we first took a bottom-up, data-driven approach to understand what visualization readers mean when they say they “trust” a visualization. We compiled and adapted a broad set of trust-related statements from existing inventories and collected responses to visualizations with varying degrees of trustworthiness. Through exploratory factor analysis, we derived an operational definition of trust in visualizations. Our findings indicate that people perceive a trustworthy visualization as one that presents credible information and is comprehensible and usable. Building on this insight, we developed an eight-item inventory: four core items measuring trust in visualizations and four optional items controlling for individual differences in baseline trust tendency. We established the inventory’s internal consistency reliability using McDonald’s omega, confirmed its content validity by demonstrating alignment with theoretically-grounded trust dimensions, and validated its criterion validity through two trust games with real-world stakes. Finally, we illustrate how this standardized inventory can be applied across diverse visualization research contexts. Utilizing our inventory, future research can examine how design choices, tasks, and domains influence trust, and how to foster appropriate trusting behavior in human-data interactions.

Abstract:
Crowd simulation plays a crucial role in various domains, including entertainment, urban planning, and safety assessment. Data-driven methods offer significant advantages in simulating natural and diverse crowd behaviors, enabling highly realistic simulations. However, existing methods often face challenges due to incomplete trajectory data and limited generalization to unfamiliar scenarios. To address these limitations, we propose a novel crowd simulation framework based on the collaboration of a small model and a large model. Inspired by the dual-process decision-making mechanism in cognitive psychology, this framework enables efficient handling of familiar scenarios while leveraging the reasoning capabilities of large models in complex or unfamiliar environments. The small model, responsible for generating fast and reactive behaviors, is trained on real-world incomplete trajectory data to learn movement patterns. The large model, which performs simulation correction to refine failed behaviors, leverages past successful and failed experiences to enhance behavior generation in complex scenarios. Experimental results demonstrate that our framework significantly improves simulation accuracy in the presence of missing trajectory segments and enhances cross-scene generalization.

Abstract:
Figurative art, as a culturally embedded medium, encodes narrative, symbolic, and emotional meanings that reflect artistic choices and historical realities. The recent availability of large-scale digital collections of figurative artworks creates opportunities for computational analysis, but existing methods mostly focus on classification or style detection, lacking structured modeling of high-level figurative elements and integration of cultural context. We present DAVA, a visual analytics system that supports interdisciplinary exploration of figurative art. First, we model paintings across three structural levels: facial expressions (micro), posture features (meso), and object co-occurrence (macro). Second, we employ a vision–language model to discover latent patterns from these features and present them through novel visualization designs. Third, we introduce domain-informed AI agents that simulate interdisciplinary research teams to interpret artworks in cultural and historical context. To evaluate DAVA, we first conducted a quantitative evaluation demonstrating the accuracy and consistency of the multi-agent interpretation mechanism. Case studies and expert interviews then confirmed the system’s utility and support for semantically and historically informed exploration of figurative art.

Abstract:
Computer-Aided Design (CAD) sketches, composed of geometric primitives and constraints, are fundamental to CAD models and play a critical role in industrial design and manufacturing. Leveraging artificial intelligence to convert hand-drawn sketches and rendered images into CAD sketches has the potential to streamline and accelerate the design process. Existing approaches predominantly focus on separately learning primitives and constraints, often employing two-stage methods or learnable tokens to model these elements independently. However, such methods fail to fully exploit the intrinsic relationships between primitives and constraints. In this paper, we propose LuBan, a lightweight, end-to-end model for CAD sketch generation that eliminates the need for separate constraint models or tokens. LuBan leverages the DEtection TRansformer (DETR) architecture for primitive modeling and distinguishes between parametric and non-parametric features. By deriving sub-primitive features from the intermediate outputs of the primitive model, LuBan facilitates constraint prediction, effectively capturing the inherent relationships between primitives and constraints. This enables the generation of CAD sketches as hierarchical graphs. Qualitative and quantitative experiments on both precise and hand-drawn renderings demonstrate that LuBan achieves state-of-the-art performance. Ablation studies further confirm its superiority over independently trained primitive models, validating its effectiveness. Additionally, LuBan embodies the principle of “what you draw is what you get,” offering significant enhancements to the design process for designers.

Abstract:
Line charts surface many features in time series data, from trends to periodicity to peaks & valleys. However, not every potentially important feature in the data may correspond to a visual feature that readers can detect or prioritize. In this study, we conducted a visual stenography task, where participants re-drew line charts to solicit information about the visual features they believed to be important. We systematically varied noise levels (SNR \approx≈ 5–30 dB) across line charts to observe how visual clutter influences which features people prioritize in their sketches. We identified three key strategies that correlated with the noise present in the stimuli: the \colorgreenReplicatorgreenReplicator attempted to retain all major features of the line chart including noise; the \coloryellowTrend KeeperyellowTrendKeeper prioritized trends disregarding periodicity and peaks; and the \colorpinkDe-noiserpinkDe-noiser filtered out noise while preserving other features. Further, we found that participants tended to faithfully retain trends and peaks & valleys when these features were present, whereas periodicity and noise were represented in more qualitative or gestural ways: semantically rather than accurately. These results suggest a need to consider more flexible and human-centric ways of presenting, summarizing, preprocessing, or clustering time series data.

Abstract:
We propose a parameter optimization method for achieving static equilibrium of discrete elastic rods. Our method simultaneously optimizes material stiffness and rest shape parameters under box constraints to exactly enforce zero net forces while avoiding stability issues and violations of physical laws. For efficiency, we split our constrained optimization problem into primal and dual subproblems via the augmented Lagrangian method, while handling the dual maximization subproblem via simple vector updates. To efficiently solve the box-constrained primal minimization subproblem, we propose a new active-set Cholesky preconditioner for variants of conjugate gradient solvers with active sets. Our method surpasses prior work in generality, robustness, and speed.

Abstract:
As virtual reality (VR) continues to evolve, traditional input methods such as handheld controllers and gesture systems often face challenges with precision, social accessibility, and user fatigue. These limitations motivate the exploration of microgestures, which promise more subtle, ergonomic, and device-free interactions. We introduce microGEXT, a lightweight microgesture-based system designed for text editing in VR without external sensors, which utilizes small, subtle hand movements to reduce physical strain compared to standard gestures. We evaluated microGEXT in three user studies. In Study 1 (N=20N=20), microGEXT reduced overall edit time and fatigue compared to a ray-casting + pinch menu baseline, the default text editing approach in commercial VR systems. Study 2 (N=20N=20) found that microGEXT performed well in short text selection tasks but was slower for longer text ranges. In Study 3 (N=10N=10), participants found microGEXT intuitive for open-ended information-gathering tasks. Across all studies, microGEXT demonstrated enhanced user experience and reduced physical effort, offering a promising alternative to traditional VR text editing techniques.

Abstract:
Augmented reality (AR) may provide supplementary information to support tasks in the physical world, offering the advantage of displaying multiple windows with high flexibility in interface layout. In AR-physical world mixed scenarios, users often need to locate and retrieve target information from virtual multi-window displays. Understanding how to design effective layouts for these interfaces is critical to enhancing visual search performance, a key element of information retrieval. This study examines the effects of depth separation, information density and curvature of virtual multi-window displays on a conjunctive visual search task in AR. Results indicate that reducing information density and introducing curvature significantly reduced both search time and the time taken to decide that the target was absent (task quit time). Although depth separation did not significantly affect search time, it notably reduced quit time. The number of errors was not significantly influenced by any of the factors. Additionally, users preferred a curved display with lower information density that remained within the device’s field of view, and their search time was fastest with this layout. Finally, we noticed variations in layout preference and performance changes among individuals, possibly influenced by differences in search strategies.

Abstract:
Mental health is an urgent societal issue, and social scientists are increasingly turning to online mental health communities (OMHCs) to analyze user behavior data for early intervention. However, existing sequence mining techniques fall short of the urgent need to explore the behavior progression of different groups (e.g., recovery or deterioration groups) and track the potential long-term impact of behaviors on mental health status. To address this issue, we introduce EMINDS, a visual analytics system built on a novel automatic mining pipeline that extracts distinct behavior stages and assesses the potential impact of frequent stage patterns on mental health status over time. The system includes a set of interactive visualizations that summarize the meaning of each behavior stage and the evolution of different stage patterns. We feature a pattern-centric Sankey diagram to reveal contextual information about the impact of stage patterns on mental health, helping experts understand the specific changes in sequences before and after a stage pattern. We evaluated the effectiveness and usability of EMINDS through two case studies and expert interviews, which examined the potential stage patterns impacting long-term mental health by analyzing user behaviors on Reddit.

Abstract:
Pseudo-haptics techniques are interesting alternatives for inducing haptic perceptions, achieved by manipulating haptic perception through the appropriate alteration of primarily visual feedback in response to body movements. However, the use of pseudo-haptics techniques with a motion-input system can sometimes be limited. This paper investigates a novel approach for extending the potential of pseudo-haptics techniques in virtual reality (VR), focusing on pseudoweight perception as the target case. The proposed approach utilizes a reaction force from force-input as a substitution of haptic cue for the pseudo-haptic perception. The paper introduced a manipulation method in which the vertical acceleration of the virtual hand is controlled by the extent of push-in of a force sensor. Such a force-input manipulation of a virtual body cannot only present pseudo-haptics with smaller physical spaces and be used by various users including physically handicapped people, but can present the reaction force proportional to the user's input to the user. We hypothesized that such a haptic force cue would contribute to the pseudo-haptic, here, the pseudoweight perception. Therefore, the paper endeavors to investigate the force-input pseudo-haptic perception in comparison with the motion-input pseudo-haptics. The paper compared force-input and motion-input manipulation in a point of achievable range and resolution of pseudo-weight. The experimental results suggest that forceinput manipulation successfully extends the range of perceptible pseudo-weight by 80% in comparison to the motion-input manipulation. On the other hand, it is revealed that motion-input manipulation has 1 step larger number of distinguishable weight levels and is easier to operate.

Abstract:
Large language models (LLMs) have achieved remarkable performance across a wide range of natural language tasks. Understanding how LLMs internally represent knowledge remains a significant challenge. Despite Sparse Autoencoders (SAEs) have emerged as a promising technique for extracting interpretable features from LLMs, SAE features do not inherently align with human-understandable concepts, making their interpretation cumbersome and labor-intensive. To bridge the gap between SAE features and human concepts, we present ConceptViz, a visual analytics system designed for exploring concepts in LLMs. ConceptViz implements a novel Identification ⇒ Interpretation ⇒Validation pipeline, enabling users to query SAEs using concepts of interest, interactively explore concept-to-feature alignments, and validate the correspondences through model behavior verification. We demonstrate the effectiveness of ConceptViz through two usage scenarios and a user study. Our results show that ConceptViz enhances interpretability research by streamlining the discovery and validation of meaningful concept representations in LLMs, ultimately aiding researchers in building more accurate mental models of LLM features. Our code and user guide are publicly available at https://github.com/Happy-Hippo209/conceptViz.

Abstract:
The interactive visual analysis of set-typed data, i.e., data with attributes that are of type set, is a rewarding area of research and applications. Valuable prior work has contributed solutions that enable the study of such data with individual set-typed dimensions. In this paper, we present CrossSet, a novel method for the joint study of two set-typed dimensions and their interplay. Based on a task analysis, we describe a new, multi-scale approach to the interactive visual exploration and analysis of such data. Two set-typed data dimensions are jointly visualized using a hierarchical matrix layout, enabling the analysis of the interactions between two set-typed attributes at several levels, in addition to the analysis of individual such dimensions. CrossSet is anchored at a compact, large-scale overview that is complemented by drill-down opportunities to study the relations between and within the set-typed dimensions, enabling an interactive visual multi-scale exploration and analysis of bivariate set-typed data. Such an interactive approach makes it possible to study single set-typed dimensions in detail, to gain an overview of the interaction and association between two such dimensions, to refine one of the dimensions to gain additional details at several levels, and to drill down to the specific interactions of individual set-elements from the set-typed dimensions. To demonstrate the effectiveness and efficiency of CrossSet, we have evaluated the new method in the context of several application scenarios.

Abstract:
In this work we study the identification of spatial correlation in distributions of 2D scalar fields, presented across different forms of visual displays. We study simple visual displays that directly show color-mapped scalar fields, namely those drawn from a distribution, and whether humans can identify strongly correlated spatial regions in these displays. In this setting, the recognition of correlation requires making judgments on a set of fields, rather than just one field. Thus, in our experimental design we compare two basic visualization designs: animation-based displays against juxtaposed views of scalar fields, along different choices of color scales. Moreover, we investigate the impacts of the distribution itself, controlling for the level of spatial correlation and discriminability in spatial scales. Our study's results illustrate the impacts of these distribution characteristics, while also highlighting how different visual displays impact the types of judgments made in assessing spatial correlation. Supplemental material is available at https://osf.io/zn4qy/

Abstract:
Interactive time-varying volume visualization is challenging due to its complex spatiotemporal features and sheer size of the dataset. Recent works transform the original discrete time-varying volumetric data into continuous Implicit Neural Representations (INR) to address the issues of compression, rendering, and super-resolution in both spatial and temporal domains. However, training the INR takes a long time to converge, especially when handling large-scale time-varying volumetric datasets. In this work, we proposed F-Hash, a novel feature-based multi-resolution Tesseract encoding architecture to greatly enhance the convergence speed compared with existing input encoding methods for modeling time-varying volumetric data. The proposed design incorporates multi-level collision-free hash functions that map dynamic 4D multi-resolution embedding grids without bucket waste, achieving high encoding capacity with compact encoding parameters. Our encoding method is agnostic to time-varying feature detection methods, making it a unified encoding solution for feature tracking and evolution visualization. Experiments show the F-Hash achieves state-of-the-art convergence speed in training various time-varying volumetric datasets for diverse features. We also proposed an adaptive ray marching algorithm to optimize the sample streaming for faster rendering of the time-varying neural representation.

Abstract:
Visualization of large-scale time-dependent simulation data is crucial for domain scientists to analyze complex phenomena, but it demands significant I/O bandwidth, storage, and computational resources. To enable effective visualization on local, low-end machines, recent advances in view synthesis techniques, such as neural radiance fields, utilize neural networks to generate novel visualizations for volumetric scenes. However, these methods focus on reconstruction quality rather than facilitating interactive visualization exploration, such as feature extraction and tracking. We introduce VolSegGS, a novel Gaussian splatting framework that supports interactive segmentation and tracking in dynamic volumetric scenes for exploratory visualization and analysis. Our approach utilizes deformable 3D Gaussians to represent a dynamic volumetric scene, allowing for real-time novel view synthesis. For accurate segmentation, we leverage the view-independent colors of Gaussians for coarse-level segmentation and refine the results with an affinity field network for fine-level segmentation. Additionally, by embedding segmentation results within the Gaussians, we ensure that their deformation enables continuous tracking of segmented regions over time. We demonstrate the effectiveness of VolSegGS with several time-varying datasets and compare our solutions against state-of-the-art methods. With the ability to interact with a dynamic scene in real time and provide flexible segmentation and tracking capabilities, VolSegGS offers a powerful solution under low computational demands. This framework unlocks exciting new possibilities for time-varying volumetric data analysis and visualization.

Abstract:
Visualization design study is a widely adopted approach for developing tailored visual solutions to domain-specific problems through close interdisciplinary collaboration. While the visualization community has proposed generalizable frameworks, there is a growing need for domain-aware methodologies that address discipline-specific challenges and refine design study practices. To investigate how domain characteristics and collaborator roles influence the design study process, we conducted interviews with 15 experts, including domain specialists from the humanities, arts, applied sciences, and artificial intelligence, as well as visualization researchers and developers, with direct experience in design studies. Our findings reveal tensions and opportunities that arise from differing expectations, communication styles, and levels of engagement among collaborators at various stages of the design process, including problem formulation, co-design, and evaluation. We highlight how domain-specific norms and role dynamics shape collaboration and influence the trajectory of visualization projects. Based on these insights, we offer practical considerations to help visualization researchers anticipate domain-specific challenges, foster mutual understanding, and adapt their methods accordingly. Our study contributes to ongoing efforts to support more context-sensitive, sustainable, and inclusive design study practices across diverse application domains.

Abstract:
Vision-language (VL) models have shown transformative potential across various critical domains due to their capability to comprehend multi-modal information. However, their performance frequently degrades under distribution shifts, making it crucial to assess and improve robustness against real-world data corruption encountered in practical applications. While advancements in VL benchmark datasets and data augmentation (DA) have contributed to robustness evaluation and improvement, there remain challenges due to a lack of in-depth comprehension of model behavior as well as the need for expertise and iterative efforts to explore data patterns. Given the achievement of visualization in explaining complex models and exploring large-scale data, understanding the impact of various data corruption on VL models aligns naturally with a visual analytics approach. To address these challenges, we introduce VisMoDAl, a visual analytics framework designed to evaluate VL model robustness against various corruption types and identify underperformed samples to guide the development of effective DA strategies. Grounded in the literature review and expert discussions, VisMoDAl supports multi-level analysis, ranging from examining performance under specific corruptions to task-driven inspection of model behavior and corresponding data slice. Unlike conventional works, VisMoDAl enables users to reason about the effects of corruption on VL models, facilitating both model behavior understanding and DA strategy formulation. The utility of our system is demonstrated through case studies and quantitative evaluations focused on corruption robustness in the image captioning task.

Abstract:
Real-time path tracing is rapidly becoming the standard for rendering in entertainment and professional applications. In scientific visualization, volume rendering plays a crucial role in helping researchers analyze and interpret complex 3D data. Recently, photorealistic rendering techniques have gained popularity in scientific visualization, yet they face significant challenges. One of the most prominent issues is slow rendering performance and high pixel variance caused by Monte Carlo integration. In this work, we introduce a novel radiance caching approach for path-traced volume rendering. Our method leverages advances in volumetric scene representation and adapts 3D Gaussian splatting to function as a multi-level, path-space radiance cache. This cache is designed to be trainable on the fly, dynamically adapting to changes in scene parameters such as lighting configurations and transfer functions. By incorporating our cache, we achieve less noisy, higher-quality images without increasing rendering costs. To evaluate our approach, we compare it against a baseline path tracer that supports uniform sampling and next-event estimation and the state-of-the-art for neural radiance caching. Through both quantitative and qualitative analyses, we demonstrate that our path-space radiance cache is a robust solution that is easy to integrate and significantly enhances the rendering quality of volumetric visualization applications while maintaining comparable computational efficiency.

Affiliations: School of Artificial Intelligence, Engineering Research Center of Intelligent Technology and Educational Application, Ministry of Education, Beijing Normal University, China; State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University, China; University of Oxford, U.K.; National Engineering Laboratory for Big Data Analysis and Application, Peking University, China; Syracuse University, USA; Center for Research on Ancient Chinese History, Peking University, China

Abstract:
Chinese calligraphy is a quintessential element of Chinese cultural heritage. Analyzing and comparing calligraphic styles not only enhances the appreciation, learning, and advancement of calligraphy but also provides valuable insights into ancient China. However, such analysis remains challenging due to the limited scalability and possible inconsistencies of qualitative methods, as well as usability and misalignment issues in conventional quantitative approaches. We propose Calli-VA, a visual analytics system, to address these challenges. Calli-VA extracts character images and their corresponding strokes from original works and characterizes each character using systematic criteria. During analysis, the system defines the analysis scope by overview and uncovers relationships between characters. Explanation and recommendation mechanisms are integrated to help users understand patterns and guide further exploration. A documentation feature allows users to record and share their findings. We demonstrate the effectiveness of Calli-VA through three case studies and expert feedback.

Abstract:
Though powerful tools for analysis and communication, interactive visualizations often fail to support real-time interaction with large datasets with millions or more records. To highlight and filter data, users indicate values or intervals of interest. Such selections may span multiple components, combine in complex ways, and require optimizations to ensure low-latency updates. We describe Mosaic Selections, a model for representing, managing, and optimizing user selections, in which one or more filter predicates are added to queries that request data for visualizations and input widgets. By analyzing both queries and selection predicates, Mosaic Selections enable automatic optimizations, including pre-aggregating data to rapidly compute selection updates. We contribute a formal description of our selection model and optimization methods, and their implementation in the open-source Mosaic architecture. Benchmark results demonstrate orders-of-magnitude latency improvements for selection-based optimizations over unoptimized queries and existing optimizers for the Vega language. The Mosaic Selection model provides infrastructure for flexible, interoperable filtering across multiple visualizations, alongside automatic optimizations to scale to millions and even billions of records.

Abstract:
With the growing availability of urban data and the increasing complexity of societal challenges, visual analytics has become essential for deriving insights into pressing real-world problems. However, analyzing such data is inherently complex and iterative, requiring expertise across multiple domains. The need to manage diverse datasets, distill intricate workflows, and integrate various analytical methods presents a high barrier to entry, especially for researchers and urban experts who lack proficiency in data management, machine learning, and visualization. Advancements in large language models offer a promising solution to lower the barriers to the construction of analytics systems by enabling users to specify intent rather than define precise computational operations. However, this shift from explicit operations to intent-based interaction introduces challenges in ensuring alignment throughout the design and development process. Without proper mechanisms, gaps can emerge between user intent, system behavior, and analytical outcomes. To address these challenges, we propose Urbanite, a framework for human-AI collaboration in urban visual analytics. Urbanite leverages a dataflow-based model that allows users to specify intent at multiple scopes, enabling interactive alignment across the specification, process, and evaluation stages of urban analytics. Based on findings from a survey to uncover challenges, Urbanite incorporates features to facilitate explainability, multi-resolution definition of tasks across dataflows, nodes, and parameters, while supporting the provenance of interactions. We demonstrate Urbanite's effectiveness through usage scenarios created in collaboration with urban experts. Urbanite is available at urbantk.org/urbanite.

Abstract:
Understanding user interactions in digital systems is essential in analyzing user behaviors and improving system usability. However, a collection of interaction sequences is often large and unstructured, making it challenging to uncover interaction patterns. To address this challenge, we introduce a visual analytics approach that integrates hierarchical clustering and process mining techniques to support analysts in exploring unstructured, large interaction sequence data. Our system employs a tailored dynamic time warping-based similarity measure to enable comparison of interaction sequences. Based on the sequence similarities, we provide stepwise, interactive navigation of clustering results with contextual visual cues for refinement and validation. We further apply process mining to characterize derived clusters. Through these hierarchical clustering and process mining steps, analysts can progressively uncover meaningful interaction patterns while utilizing visual guidance and incorporating domain expertise. We demonstrate our system's effectiveness and applicability through two case studies involving system designers, developers, and domain experts.

Abstract:
Biomedical data harmonization is essential for enabling exploratory analyses and meta-studies, but the process of schema matching-identifying semantic correspondences between elements of disparate datasets (schemas)-remains a labor-intensive and error-prone task. Even state-of-the-art automated methods often yield low accuracy when applied to biomedical schemas due to the large number of attributes and nuanced semantic differences between them. We present BDIViz, a novel visual analytics system designed to streamline the schema matching process for biomedical data. Through formative studies with domain experts, we identified key requirements for an effective solution and developed interactive visualization techniques that address both scalability challenges and semantic ambiguity. BDIViz employs an ensemble approach that combines multiple matching methods with LLM-based validation, summarizes matches through interactive heatmaps, and provides coordinated views that enable users to quickly compare attributes and their values. Our method-agnostic design allows the system to integrate various schema matching algorithms and adapt to application-specific needs. Through two biomedical case studies and a within-subject user study with domain experts, we demonstrate that BDIViz significantly improves matching accuracy while reducing cognitive load and curation time compared to baseline approaches.

Abstract:
Eligibility criteria play a critical role in clinical trials by determining the target patient population, which significantly influences the outcomes of medical interventions. However, current approaches for designing eligibility criteria have limitations to support interactive exploration of the large space of eligibility criteria. They also ignore incorporating detailed characteristics from the original electronic health record (EHR) data for criteria refinement. To address these limitations, we proposed TrialCompass, a visual analytics system integrating a novel workflow, which can empower clinicians to iteratively explore the vast space of eligibility criteria through knowledge-driven and outcome-driven approaches. TrialCompass supports history-tracking to help clinicians trace the evolution of their adjustments and decisions when exploring various forms of data (i.e., eligibility criteria, outcome metrics, and detailed characteristics of original EHR data) through these two approaches. This feature can help clinicians comprehend the impact of eligibility criteria on outcome metrics and patient characteristics, which facilitates systematic refinement of eligibility criteria. Using a real-world dataset, we demonstrated the effectiveness of TrialCompass in providing insights into designing eligibility criteria for septic shock and sepsis-associated acute kidney injury. We also discussed the research prospects of applying visual analytics to clinical trials.

Abstract:
Political sectarianism is fueled in part by misperceptions of political opponents: People commonly overestimate the support for extreme policies among members of the other party. These misperceptions inflame partisan animosity and may be used to justify extremism among one's own party. Research suggests that correcting partisan misperceptions-by informing people about the actual views of outparty members-may reduce one's own expressed support for political extremism, including partisan violence and antidemocratic actions. However, there remains a limited understanding of how the design of correction interventions drives these effects. The present study investigated how correction effects depend on different representations of outparty views communicated through data visualizations. Building on prior interventions that present the average outparty view, we consider the impact of visualizations that more fully convey the range of views among outparty members. We conducted an experiment with U.S.-based participants from Prolific (N=239 Democrats, N=244 Republicans). Participants made predictions about support for political violence and undemocratic practices among members of their political outparty. They were then presented with data from an earlier survey on the actual views of outparty members. Some participants viewed only the average response (Mean-Only condition), while other groups were shown visual representations of the range of views from 75% of the outparty (Mean+Interval condition) or the full distribution of responses (Mean+Points condition). Compared to a control group that was not informed about outparty views, we observed the strongest correction effects (i.e., lower support for political violence and undemocratic practices) among participants in the Mean-only and Mean+Points condition, while correction effects were weaker in the Mean+Interval condition. In addition, participants who observed the full distribution of out-party views (Mean+Points condition) were most accurate at later recalling the degree of support among the outparty. Our findings suggest that data visualizations can be an important tool for correcting pervasive distortions in beliefs about other groups. However, the way in which variability in outparty views is visualized can significantly shape how people interpret and respond to corrective information. Supplemental materials for this paper are available at this OSF repository.

Abstract:
Interpreting data visualizations is an essential skill in today's education, yet students often struggle with understanding unfamiliar formats. This study investigates how four learning materials - textbook, comic, video, and game - affect middle- and high school students' ability to interpret line charts, area charts, stacked area charts, and stream graphs. We conducted a comparative classroom study with 68 students, using pre- and post-tests, worksheet activities, and group discussions to assess learning outcomes and understanding. Our results show statistically significant improvement in students' understanding of stacked area charts and stream graphs, while no significant differences between the learning materials were found. This suggests that more factors than initially anticipated - such as engagement, motivation and active learning strategies - influence the learning outcome. The analysis of the worksheets revealed that while students could infer surface-level insights from charts, over 70% struggled to identify underlying patterns or relationships. Additionally, a common challenge across all learning materials was reading fatigue, which often led students to skim content, disengage, or misinterpret key information. These findings highlight the need for educational tools and approaches that foster deeper understanding of unfamiliar visualizations, reduce cognitive load, and encourage active engagement.

Abstract:
In this paper, we study the propagation of data uncertainty through the marching cubes algorithm for isosurface visualization for correlated uncertain data. Consideration of correlation has been shown paramount for avoiding errors in uncertainty quantification and visualization in multiple prior studies. Although the problem of isosurface uncertainty with spatial data correlation has been previously addressed, there are two major limitations to prior treatments. First, there are no analytical formulations for uncertainty quantification of isosurfaces when the data uncertainty is characterized by a Gaussian distribution with spatial correlation. Second, as a consequence of the lack of analytical formulations, existing techniques resort to a Monte Carlo sampling approach, which is expensive and difficult to integrate into visualization tools. To address these limitations, we present a closed-form framework to efficiently derive uncertainty in marching cubes level-sets for Gaussian uncertain data with spatial correlation (MAGIC). To derive closed-form solutions, we leverage the Hinkley’s derivation on the ratio of Gaussian distributions. With our analytical framework, we achieve a significant speed-up and enhanced accuracy of uncertainty quantification over classical Monte Carlo methods. We further accelerate our analytical solutions using many-core processors to achieve speed-ups up to \text585 ×585× and integrability with production visualization tools for broader impact. We demonstrate the effectiveness of our correlation-aware uncertainty framework through experiments on meteorology, urban flow, and astrophysics simulation datasets.

Abstract:
Domain-specific visualizations sometimes focus on narrow, albeit important, tasks for one group of users. This focus limits the utility of a visualization to other groups working with the same data. While tasks elicited from other groups can present a design pitfall if not disambiguated, they also present a design opportunity—namely, the development of visualizations that support multiple groups. This development choice presents a trade-off of broadening the scope but limiting support for the more narrow tasks of any one group, which in some cases can enhance the overall utility of the visualization. We investigate this scenario through a design study where we develop Guidepost, a notebook-embedded visualization of data that helps scientists assess compute wait times, machine learning researchers understand prediction accuracy, and system maintainers analyze usage trends. We adapt the use of personas for visualization design from existing literature in the HCI and design domains, applying them to categorize tasks based on their uniqueness across stakeholder personas. Under this model, tasks shared between all groups should be supported by interactive visualizations and tasks unique to each group can be deferred to scripting with notebook-embedded visualization design. We evaluate our visualization through real-world case studies and a task-focused evaluation with nine participants. We observe that together, Guidepost’s visual encodings, interactions, and export capabilities support the tasks of our differing personas.

Abstract:
This study investigates how partner avatar design affects learning and memory when an avatar serves as a lecturer. Based on earlier research on the environmental context dependency of memory, we hypothesize that the use of diverse partner avatars results in a slower learning rate but better memory retention than that of a constant partner avatar. Accordingly, participants were tasked with memorizing Tagalog–Japanese word pairs. On the first day of the experiment, they repeatedly learned the pairs over six sessions from a partner avatar in an immersive virtual environment. One week later, on the second day of the experiment, they underwent a recall test in a real environment. We employed a between-participants design to compare the following conditions: the varied avatar condition, in which each repetition used a different avatar, and the constant avatar condition, in which the same avatar was used throughout the experiment. Results showed that participants in the varied avatar condition recalled significantly worse during the learning trials on the first day. However, we found no significant difference between conditions in the delayed recall test on the second day. We discuss these effects in relation to the social presence of the partner avatar. This study opens up a novel approach to optimizing the effectiveness of instructor avatars in immersive virtual environments.

Abstract:
Group navigation is a virtual reality (VR) technology designed to replace personal navigation, enhancing efficiency and aiding guided tours in path planning for other users. Group navigation techniques require users to have a clear understanding of navigation and travel modalities. Tour guides must ensure that users do not intersect with the environment or other users’ models and must provide reasonable tour formations. To address these needs, we propose a scene-aware automatic formation generation method for fast and easy-to-use guided tours. First, we generate a limited number of candidate points for the exhibition and initialize the visiting group formation. Next, we optimize the formation to maximize view quality using our proposed viewpoint observation score. Finally, we match the visitors to the optimized formation to ensure a minimal view deflection angle. Moreover, we conducted a user study to evaluate the performance of our approach. Compared to the current method, our approach significantly increased navigation efficiency and view quality. Additionally, it substantially decreased the task load and improved the system usability for both tour guides and visitors.

Abstract:
We propose Transmission-Reflection Gaussians (TR-Gaussians), a novel 3D-Gaussian-based representation for high-fidelity rendering of planar transmission and reflection, which are ubiquitous in indoor scenes. Our method combines 3D Gaussians with learnable reflection planes that explicitly model the glass planes with view-dependent reflectance strengths. Real scenes and transmission components are modeled by 3D Gaussians and the reflection components are modeled by the mirrored Gaussians with respect to the reflection plane. The transmission and reflection components are blended according to a Fresnel-based, view-dependent weighting scheme, allowing for faithful synthesis of complex appearance effects under varying viewpoints. To effectively optimize TR-Gaussians, we develop a multi-stage optimization framework incorporating color and geometry constraints and an opacity perturbation mechanism. Experiments on different datasets demonstrate that TR-Gaussians achieve real-time, high-fidelity novel view synthesis in scenes with planar transmission and reflection, and outperform state-of-the-art approaches both quantitatively and qualitatively.

Abstract:
While at first glance one might assume haptic devices for Virtual Reality (VR) would directly translate to use in Augmented Reality (AR), key differences in how users interact with these environments create distinct design constraints. AR allows users to see their surroundings and interact with both physical and virtual elements, requiring virtual objects to convey the same sense of weight as their physical counterparts. However, force feedback solutions developed for the more mature field of VR are not compatible with this blended environment as they occlude the user’s hand and prevent interaction with physical objects. We present a novel exoskeleton, Wriality, that utilizes a minimal feedback design to render weight-related forces while maintaining natural engagement with an AR environment, providing a more immersive experience than current state-of-the-art solutions. We present the design and characterization of Wriality and then validate its effectiveness with a human subject experiment. We show that, with force compensation, kinesthetic feedback alone can allow a virtual object to be perceived as equal in weight to its real twin. We show that this force compensation varies with both grip method and with the weight of an object. We then present the subjective feedback on user experience, specifically comfort level, and the effects of the device on realism and immersiveness in AR. Our findings show that the minimal force feedback approach exemplified in Wriality is still effective in providing high levels of immersion by incorporating a force compensation scheme based on intended grip and the object’s weight. The results of our user feedback also inform design considerations for future haptic devices targeting AR.

Abstract:
Fluid simulation is a central topic in computer graphics, encompassing a wide range of methodologies for modeling Newtonian, non-Newtonian, and viscoelastic behaviors across both single-phase and multiphase settings. Existing single-phase frameworks have achieved high visual fidelity, yet multiphase simulations remain limited in accurately capturing complex phase interactions, particularly under high-viscosity-ratio or viscoelastic conditions. To address these challenges, we develop a unified multiphase viscoelastic formulation capable of handling diverse fluid types—including Newtonian, shear-dependent non-Newtonian, and viscoelastic flows—within a single consistent framework. The formulation extends mixture-model approaches through a multi-mode conformation tensor representation, which enhances numerical stability via phase-level stress corrections and efficiently captures a broad spectrum of rheological behaviors. Compared with existing techniques, our framework achieves improved momentum–mass consistency and numerical stability, maintaining physically plausible results across wide viscosity ranges, advancing the state of the art in multiphase viscoelastic fluid simulation.

Abstract:
Surfacing peers' workload and speed can reshape collaboration in VR, yet prior work often conflates social cues with environmental load. We isolate the social channel with TRACE-VR, which holds geometry, physics, rules, and timing constant while independently manipulating effort identifiability (traceable vs. anonymous) and peer effort (pace) (high vs. low). In a 2×2 within-subjects study (n=32), each participant worked with nine scripted co-actors for three minutes, transporting 16 crates along a self-chosen path between fixed pickup/drop points; co-actors followed fixed pre-authored routes. We measured intrinsic motivation, social-evaluative load, NASA-TLX, and behavioral/process outcomes. Our findings show that identifiability and higher peer effort (pace) each modestly increased completion rates, with asymmetric effects suggesting partial substitution between accountability cues and normative pace. Perceived monitoring and temporal demand depended on their combination—identifiability raised monitoring at low pace and amplified time pressure at high pace—while composite TLX and finishers' times remained comparable across conditions. Thus, these cues chiefly determine who finishes rather than how fast finishers move. Motivation did not uniformly increase under higher pace. We frame identifiability and pace as partially substitutable levers for shaping social-evaluative experience and facet-level motivation, and outline tempo-aware, accountability-aware guidance for collaborative VR.

Abstract:
We propose a method for generating task breakpoints based on an Origin-Centric Graph (OCG) to segment goal-oriented activity recordings into task units for adaptive playback in Virtual Reality (VR) environments. With the development of Augmented Reality (AR)/VR head-mounted displays (HMDs), research on adaptive tutorials and authoring tools has become active, but existing task segmentation methods mainly rely on manual annotation or are restricted to 2D video which limits their applicability to 3D VR contexts. In our approach, assembly scenarios with clearly defined task boundaries are recorded using a structured spatio-temporal scene graph (STSG), and the OCG is employed to track changes in the central object and the formation of new groups, thereby generating task breakpoints automatically. A user study collected user-perceived task breakpoints to establish ground truth (GT), and comparison with the algorithm-detected breakpoints demonstrated high agreement and confirmed accuracy in supporting adaptive playback. The proposed task segmentation method provides a foundation for dynamically adjusting VR playback according to user proficiency and progress, with potential for extension into automatic timeline segmentation systems for diverse VR recordings.

Abstract:
Social Virtual Reality (SVR) enables users to embody avatars and interact with others in shared virtual environments. On platforms such as Rec Room, VRChat, and Meta Horizon Worlds, inappropriate social behaviours are common and can lead to harmful emotional experiences for users. To prevent or mitigate these negative effects, SVR platforms offer a range of actions, referred to as safety tools, that allow users to control the information and interactions they are exposed to. Although prior research has analysed and proposed initial classifications of these tools, a structured, theory-driven approach to their characterization is still lacking. In this paper, we propose a taxonomy of safety tools for SVR platforms inspired by existing literature in social psychology and human-computer interaction. In order to characterise these tools, we focus on the area of social interaction, which occurs between a sender and a receiver, within a social context of the SVR medium. Our approach not only provides a model of social interaction in SVR, but also describes how safety tools function, including the processes of selecting and manipulating them, factors that are critical to their effectiveness. Finally, we apply our framework to characterise existing safety tools, identify their limitations, and present design guidelines for future SVR platforms.

Abstract:
Visualization of complex architectural and interior design proposals demands spatial perspective taking (SPT), an ability that can be difficult to acquire. Mixed Reality (MR) is particularly effective in architectural visualization because it overlays virtual models onto already existing real environments, allowing for precise spatial accuracy and aesthetic assessment. However, MR confines users to viewpoints tied to locations they can physically reach in the real-world. To overcome this constraint, we designed an applied, visibility-driven SPT task and explored two methods to provide additional perspective cues in MR: the World-in-Miniature (WIM) mode and the Cross Reality (CR) mode. The WIM mode provides a scaled-down 3D replica of the virtual environment, allowing users to explore a design at a reduced scale. Conversely, the CR mode allows switching from MR to VR to hide the real-world completely, enabling users to teleport to physically unreachable viewpoints and assess a design at an accurate scale. We conducted a pilot user study with 14 participants and a formal user study with 80 participants on an object-placement task. Our results revealed that the CR mode achieved success scores as high as those of the WIM mode. The WIM mode was faster and perceived as more user-friendly. However, participants preferred the CR mode for its accurate spatial representation and effective presentation of design ideas. The CR mode recreates the physical room as a scale-accurate virtual counterpart, preserving geometric cues critical for spatial visualization while removing the need for manual room modeling. Our findings highlight the potential of these modes to transform 3D design visualization and set new standards in the industry.

Abstract:
Multi-user virtual reality (VR) systems support multiple locomotion methods, but coordination may be challenging when partners use different methods. We studied dyads navigating a shared virtual maze using symmetric (same) or asymmetric (different) locomotion methods—specifically, steering (continuous movement with physical turning) and teleportation (point-and-teleport within 30 meters with physical turning)—and assessed survey knowledge using direction and straightline distance estimates between objects. We found that (1) dyads acquired better survey knowledge than individuals across locomotion conditions; (2) teleportation showed no significant difference from steering in distance error and yielded significantly lower angular error; (3) asymmetric locomotion methods did not significantly increase cybersickness compared to symmetric locomotion methods; and (4) symmetric and asymmetric dyads showed no significant differences in distance or angular errors. Overall, these results suggest that collaborative navigation improves spatial learning and that asymmetric locomotion can improve accessibility without significantly increasing cybersickness in multi-user VR applications.

Abstract:
A novel projective device, the AR Magic Lantern (ARML), has allowed Spatial Augmented Reality (SAR) to be deployed as a portable AR experience comparable to those on other technologies such as smartphones and headsets. In this study we compare user experience in a real virtual heritage application between two technologies that are growing in feasibility and popularity: SAR (using the ARML) and passthrough headset-AR (using the Meta Quest 3). Given that managers and curators of heritage sites want their visitors to attain a good awareness of the physical site, AR should provide a good blend between the physical and the augmented, without technology getting in the way. The results obtained seem to indicate that SAR provides higher awareness of the physical site than headset-AR. Moreover, the SAR seems to spontaneously spark experience exchange between the user and other visitors, while headset-AR tends to isolate users. Finally, headset-AR seems to put users in a task-oriented mindset while SAR seems to better contextualize the experience and its content within the physical site.

Abstract:
Redirected walking (RDW) allows people to explore large virtual environments while walking within a smaller physical space. When physical space is limited, explicit resets such as turn-in-place are unavoidable. Previous studies have reduced the frequency of resets by adjusting user paths. However, resets remain necessary under severe spatial constraints. The frequent use of turn-in-place resets interrupts locomotion and can degrade task performance and user experience. We present Non-Euclidean Portal Reset (NEPR), a reset technique that enables continuous experiences without pausing the user's walking. When a collision risk is detected, NEPR opens a virtual portal leading to a short, non-Euclidean corridor. Traversing the corridor repositions and reorients the user. The exit returns the user near the point of interest target in the primary world, maintaining flow. To evaluate the effectiveness of NEPR, we conducted user experiments comparing (1) the conventional turn-in-place reset, (2) NEPR, and (3) a hybrid method combining both approaches. Our results demonstrate that NEPR and the combined technique significantly improve the user experience and task performance compared to the traditional method. Overall, NEPR reframes resets as seamless transitions rather than interruptions, enhancing the practicality of RDW.

Abstract:
This paper presents an approach for extending 360 VR to support viewpoint translation and collaboration, without increasing the acquisition cost. The approach models the ground under the 360° image acquisition viewpoint with a planar patch, on which the user stands and walks. The planar patch is connected to the background with visual continuity. The approach anchors the user's avatar to the ground patch, and supports collaboration, with multiple avatars standing on the patch. The approach only requires defining two trivial parameters: the approximate acquisition height, and the size of the patch. Therefore, the approach improves the immersive exploration of a real-world scene captured with a 360° image, without increasing the cost of the acquisition of the virtual environment. The approach was evaluated in a controlled user study (N = 30) with five user t asks. The results show that the approach has an advantage over conventional 360 VR in terms of anchoring the user to the ground, of user viewpoint translation, of collaboration, and of accurate integration of stationary and dynamic virtual objects into the 3D scene.

Abstract:
Behavioral authentication has become increasingly popular as a natural method for authentication in Virtual Reality (VR). However, existing studies often overlook the fact that users may perform behavioral authentication in different postures (i.e., sitting, standing, reclining) during VR use. Therefore, understanding how posture variations affect classification accuracy is crucial for designing posture-robust systems. In this study, we conducted a controlled experiment (N = 30) to investigate the impact of posture on classification accuracy during a target-selection task. We collected behavioral trajectory data and analyzed it using multivariate time series classification algorithms, addressing authentication performance under three different postures. In a within-posture authentication, reclining took longer but achieved the highest classification accuracy, with an interaction effect between posture and target vertical layout. In cross-posture authentication, transfers from sitting to standing/reclining were more effective than direct transfers between standing and reclining, with vertical layout crucial for classification accuracy. In mixed-posture training, the cross-posture classification accuracy increased, particularly when standing and reclining data were combined to help the model indirectly learn features of sitting posture. These findings provide valuable insights for designing tasks and data collection strategies that support the development of robust cross-posture authentication systems.

Abstract:
The efficient utilization of underground spaces is a crucial strategy for mitigating land scarcity and expanding habitable environments. 3D Gaussian Splatting (3DGS) has emerged as a key enabler for enhancing robotic perception and spatial digitalization in underground spaces due to its unique advantages. However, illumination variation, sensor noise, and geometric degradation of underground spaces significantly degrade the localization accuracy, which compromises the robustness of existing Simultaneous Localization and Mapping (SLAM) systems. Therefore, we propose an RGB-D perception-enhanced 3DGS SLAM method. First, a multi-dimensional data enhancement and correction pipeline is introduced, integrating Multi-Scale Retinex with Color Restoration (MSRCR), Side Window Filtering (SWF), and adaptive gamma correction in the Hue-Intensity-Saturation (HIS) color space to improve low-light visual fidelity, while a depth completion network is developed to perform hole-filling in depth data. Second, a keyframe selection method based on the hybrid metric is proposed, which incorporates consistency constraints and multi-view overlap analysis to balance computational efficiency and representational completeness. Finally, a dual-constraint Gaussian management strategy is introduced, integrating an opacity threshold and observation frequency to filter out invalid Gaussian ellipsoids. At the same time, loop closure detection ensures global trajectory consistency and mapping accuracy. To validate the proposed method, experiments were conducted in typical underground spaces, including coal mine tunnels and underground parking garages, using a custom-designed underground mobile robot platform. The results demonstrate that, compared to state-of-the-art methods, the proposed method achieves a 13.8% improvement in Peak Signal-to-Noise Ratio (PSNR) over the best benchmark, while also achieving competitive trajectory accuracy and computational efficiency. These findings provide strong support for the development of digital twin systems in underground spaces.

Affiliations: Biomedical Imaging Group Rotterdam, Department of Radiology & Nuclear Medicine, Erasmus MC, Rotterdam, GD, The Netherlands; Department of Electrical Engineering, Biomedical Diagnostics, Eindhoven University of Technology, Eindhoven, AZ, The Netherlands; Department of High Performance Compute & Visualization, SURF bv, Utrecht, EP, The Netherlands; Department of Neurosurgery, Erasmus MC, Rotterdam, GD, The Netherlands; Trauma Research Unit, Department of Surgery, Erasmus MC, Rotterdam, GD, The Netherlands; Department of Oral and Maxillofacial Surgery, Erasmus MC, Rotterdam, GD, The Netherlands

Abstract:
Augmented reality (AR) devices with head mounted displays (HMDs) facilitate direct superimposition of 3D preoperative imaging data onto the patient during surgery. To use an HMD-AR device as a stand-alone surgical navigation system, the device should be able to locate the patient and surgical instruments, align preoperative imaging data with the patient, and visualize navigation data during surgery. Whereas some of the technologies required for this are known, integration in such devices is cumbersome and requires specific knowledge and expertise, hampering scientific progress in this field. This work therefore aims to present and evaluate an integrated HMD-based AR surgical navigation framework that is adaptable to diverse surgical applications. The framework tracks 2D patterns as reference markers attached to the patient and surgical instruments. It allows for the calibration of surgical tools using pivot and reference-based calibration techniques. It enables image-to-patient registration using point-based matching and manual positioning. The integrated functionalities of the framework are evaluated on two HMD devices, the HoloLens 2 and Magic Leap 2, with two surgical use cases being evaluated in a phantom setup: AR-guided needle insertion and rib fracture localization. The framework was able to achieve a mean tooltip calibration accuracy of 1 mm, a registration accuracy of 3 mm, and a targeting accuracy below 5 mm on the two surgical use cases. The framework presents an easy-to-use configurable tool for HMD-based AR surgical navigation, which can be extended and adapted to many surgical applications.

Abstract:
In the era of Big Data, augmented reality (AR)-based 3D visualization technology is gradually becoming an essential tool of effective information dissemination. However, 3D data visualization inevitably faces a series of challenges, particularly when visualizing large datasets within limited physical spaces. This often results in high spatial density, characterized by smaller target objects and occlusion of data objects, which greatly complicate object selection and subsequent interaction. To address these issues, this paper proposes a multimodal progressive target selection technique—StarPicker, which integrates the SpotLight metaphor and Clock metaphor. By leveraging wrist rotation and gesture interactions, StarPicker facilitates a coarse-to-fine semantic disambiguation and precise selection of target objects. To validate the effectiveness of StarPicker, a comparative user study was conducted against the state-of-the-art techniques (GridWall and FlowerCone) under high-density (up to 360 objects) and small-target (1cm) conditions. Experimental results demonstrate that StarPicker significantly outperforms the baselines in terms of target selection accuracy and user satisfaction, while achieving comparable or better completion times, especially in denser scenes. This work offers a novel approach to target selection and interaction technology in the field of AR-based 3D Big Data visualization.

Abstract:
Extended Reality (XR) shows great potential in the manufacturing industry. It can help designers make initial proposals and iterations at low cost before manufacturers and investors decide whether to invest in research, development or production. According to the literature, prototyping in XR is easier to use than three-dimensional (3D) modelling with a personal computer and more capable of displaying 3D structures than paper drawing. In this comprehensive review, we systematically surveyed 68 manuscripts that met our search criteria on prototyping in XR, and discussed the possibility of transferring created virtual prototypes from XR to physical fabrication. We addressed five research questions about the main elements, display devices, user interaction methods, fabrication links, and future challenges. Based on these questions, we identified a clear technology trend with Head-Mounted Displays (HMDs) being used in more than half the studies (53%). Our analysis also shows that nearly half the literature (32 of the 68 articles) focus on ‘Model Construction and Rendering’, whereas the crucial ‘Link to Fabrication’ is addressed in only 10% of studies, highlighting a critical gap. In view of these latest developments and gaps, we speculate on the challenges and opportunities in the field of prototyping in XR, especially in linking extended reality to fabrication, with the aim of guiding researchers towards new research directions.

Abstract:
This study explores a streamlined facial data collection method for conversational contexts, addressing the limitations of existing approaches that often require extensive datasets and prioritize technical metrics over user perception and experience. We systematically investigate which facial expression data are essential for reconstructing photorealistic avatars and how they can be captured efficiently. Our research employs a two-phase methodology to identify efficient facial data collection strategies and evaluate their effectiveness. In the first phase, we conduct facial data acquisition and evaluate reconstruction performance using utterance data and emotional data. In the second phase, we carry out a comprehensive user evaluation comparing three progressive conditions: utterance only, utterance and emotional data, and a control condition involving extensive data. Findings from 24 participants engaged in simulated face-to-face conversations reveal that targeted utterance and emotional data achieve comparable levels of perceived realism, naturalness, and telepresence, while reducing training time and data usage when compared to the extensive data collection approach. These results demonstrate that targeted data inputs can enable efficient avatar face reconstruction, offering practical guidelines for real-time applications such as AR/VR telepresence and highlighting the trade-off between data quantity and perceived quality.

Abstract:
Redirected Walking (RDW) enables navigation in large virtual worlds by manipulating a user's viewpoint, yet the perceptual thresholds governing these manipulations are often treated as fixed sensory limits. In this work, we propose the concept of Semantic Anchors: objects that stabilise self-motion perception through high-level cognitive priors. In a psychophysical experiment (N=22), participants judged virtual rotations while viewing familiar or arbitrary objects of varying sizes and rotational reference frames. Results reveal that scene semantics do not merely modulate thresholds but fundamentally alter the integration of spatial cues. When a familiar object's canonical size was violated to create a specific size-congruency conflict, the window for imperceptible redirection expanded significantly, allowing rotation gains of up to 30% to remain imperceptible. Crucially, familiarity inverted the geometric dependence observed with arbitrary shapes: while geometric frames dictated sensitivity for abstract objects, semantic stability dominated for familiar ones. These findings demonstrate that RDW thresholds are dynamically co-constructed by sensory input and scene plausibility, offering a new principle for designing semantically informed locomotion systems that leverage cognitive expectations to expand explorable space.

Abstract:
Virtual reality (VR) is compelling precisely because it transcends physical realism, allowing reach to be stretched, motion rescaled, and objects relocated. While these distortions expand interaction possibilities, they lack principled limits: How far can they go before agency, ownership, or presence collapse? We present a dual-axis framework that decomposes distortions into metric deviations (continuous remappings akin to stretching) and topological breaks (discontinuous jumps akin to cutting), both applied to the same hand-target relation. To operationalize this framework, we introduce the Single-Interaction Dual-Axis (SIDA) protocol, combining a standardized task with psychophysical estimation of Just-Acceptable Distortion (JAD) and Breaks in Presence (BiP) under stable or jittered mappings. In a study with N=52, results show that metric deviations are significantly more tolerable than topological breaks, predictability expands tolerance by approximately 35%, and combined distortions interact contractively, constraining the total budget. Our contributions include (i) a unified model for commensurate comparison, (ii) a reproducible protocol for threshold estimation, and (iii) the comfort envelope, a multi-dimensional budget of acceptable distortion that guides trade-offs between agency and presence in VR design.

Abstract:
Intergenerational households often involve children and older adults sharing daily life. Yet, it remains challenging to find shared exercise activities due to the significant differences in motor ability, motivational preferences, and user experience needs. Existing exergames lower the entry barriers through body-based interaction, but screen-based setups restrict play to indoor environments. XR HMDs extend exergames into public contexts, yet the differential impact of VR and MR on intergenerational play remains underexplored. To address this gap, we developed an asymmetric-role XR exergame and conducted a 2× 2 within-subjects study with 20 intergenerational pairs (N=40) to explore the effect of Display Mode (VR vs MR) and Environmental Context (Private Space vs Public Space) on their user experience and collaborative performance. This work reveals clear age-related patterns: children exhibited greater engagement in immersive VR environments, whereas older adults preferred MR for improved safety and enhanced visibility of their intergenerational partners. Finally, we offer design guidelines for leveraging complementary strengths of XR display modes and environmental contexts to support inclusive and engaging intergenerational exergames.

Abstract:
Teleportation is a widely adopted locomotion technique in virtual reality (VR), with reorientation commonly integrated into commercial systems to reduce physical turning and improve navigational control. However, research on reorientation-enabled teleportation has received limited attention and still lacks clear design guidance, resulting in longer interaction times, greater complexity, and higher mental workload. To bridge this gap, we revisit reorientation-enabled teleportation through bare-hand interaction, a natural and device-free modality that is now increasingly supported and widely adopted in mainstream VR systems. Accordingly, four gesture-based techniques, Concurrent Wrist Rotation (C-WR), Decoupled Wrist Rotation (D-WR), Joystick Wrist Rotation (J-WR), and Direction Drawing (DD), were introduced and evaluated in a controlled user study with 24 participants. Results show that C-WR was fastest but error-prone, D-WR improved accuracy at the cost of time, J-WR supported precise orientation but was demanding and slow, while DD offered balanced performance and was most preferred overall. From these findings, four design implications were distilled, emphasizing efficiency, intuitive mappings, and support for diverse user preferences. Finally, two example extensions illustrate how these techniques may extend to broader VR applications.

Abstract:
Human Motion Simulation (HMS) predicts and visualizes how virtual agents move and behave in various scenarios, aiding decision-making. However, it does not communicate the experience behind agents' behavior. To address this, we present Embodied Spatial Simulation (ESS), a concept combining augmented reality(AR) and HMS to support in-situ decision-making. It enables an XR user to create “what-if” scenarios, simulate virtual agents' behavior in situ, and experience the scenarios from their perspectives through embodiment. To demonstrate ESS, we developed a proof-of-concept prototype featuring two scenarios (e.g., fire evacuation planning). From expert walkthroughs (N=4) and a survey (N=38) with different professions, we found that incorporating embodiment into simulations can change users' intentions to use simulation outcomes. Rather than directly obtaining solutions from the results, users showed more interest in understanding the experience behind simulated behavior and using this insight to shape decisions. We discuss the potential of ESS to support future human-centered design practices.

Abstract:
Mixed Reality (MR) environments are reshaping education by fostering engagement and motivation. Within such environments, Intelligent Virtual Agents (IVAs) can serve as personalized guides, tutors, or collaborative partners. In this research, we present a novel IVA for education that perceives the learner's environment, adapts its guidance to their progress, and interacts with both the learner and the real world. Our multimodal IVA integrates speech capabilities powered by large language models (LLMs) with computer vision for dynamic scene understanding, allowing it to converse naturally with learners and provide adaptive support based on their progress. To evaluate our IVA, we implemented a Scratch programming scenario using the Magic Leap 2 headset. We conducted a user study with 24 university students (12 male, and 12 female) who completed the same tasks using three systems: basic AR cues, a chatbot with AR cues, and our IVA. Quantitative data were obtained using standardized questionnaires such as the Technology Acceptance Model (TAM), the User Engagement Scale (UES), and the System Usability Scale (SUS), alongside qualitative feedback. The results revealed that although the IVA system took a longer time compared to other systems, it fostered greater motivation and engagement, achieved the highest SUS score, and showed the strongest TAM correlations. Participants perceived IVA as the most valuable system, highlighting the potential of multimodal embodied agents as virtual tutors in MR.

Abstract:
Augmented Reality can induce illusory perceptions of heat when users experience virtual flames rendered on their hands. In this exploratory study, we used high-density electroencephalography to investigate the neural correlates associated with this experience. We recorded neural activity in a within-subjects design during three key audiovisual conditions: (A) lifelike virtual flames on the hand paired with a burning sound; (B) a non-fire-like control stimulus (green, downward-flowing visuals with a synthetic sound) on the hand; and (C) realistic flames presented between the hands. Our preliminary analyses revealed distinct electroencephalography signatures between the Fire-on-Hand condition and both control conditions. Our findings suggest that AR-induced thermal illusions may engage brain networks involved in somatosensation, multisensory body representation, and spatial attention. This top-down, predictive mechanism appears to recruit neural systems that overlap with those involved in the higher-order cortical processing of actual thermal stimuli.

Abstract:
This study investigates how interpersonal synchrony (IPS) in full-body interactive Mixed Reality (MR) environments can shape children's prosocial behavior. We introduce DragonIce 2.0, a two-part MR game designed to induce synchronous or asynchronous group activity and assess their impact on prosocial behavior. To examine changes in affiliation (team belonging and peer liking) and collaboration (overt joint actions), we conducted a comparative study with children aged 9–10, using both system logs and self-report measures. The results highlight the importance of context, task structure, and perceived coordination in shaping social outcomes, offering insights into how interaction dynamics shape social engagement in immersive settings. This work contributes a novel approach for studying prosociality in ecologically valid, technology-mediated settings by combining playful, immersive interaction with controlled manipulation of group synchrony. Our findings inform the design of inclusive MR experiences that support social connection, collaboration, and group engagement among children.

Abstract:
The N-back task is widely recognized as a paradigm for cognitive training due to its adaptability and effectiveness in mitigating working memory decline. However, its repetitive and monotonous design often leads to reduced engagement and poor long-term adherence, lowering intervention effectiveness. To address these limitations, researchers have turned to game-based interventions and immersive technologies to enhance user motivation and sustain training participation. In this study, we propose VRN-back, a music rhythm-based gamified cognitive training system in virtual reality (VR). The system combines the adaptive N-back paradigm with rhythm-game mechanics to increase enjoyment and motivation, while immersive interaction fosters engagement and enhances training outcomes. We conducted a seven-day study with 15 young adults to evaluate the system's impact on their working memory, transfer to related cognitive domains, and user experience. Results showed significant improvements in N-back accuracy, reaction time, and maximum N level, as well as transfer effects on attention and inhibitory control tasks. Subjective evaluations indicated good usability and user experience, with the System Usability Scale scoring 80.83/100, the short User Experience Questionnaire rating pragmatic quality as “Good” and hedonic quality as “Above Average”, and the Pleasure-Arousal-Dominance scale confirming sustained positive emotional states. Simulator sickness remained minimal before and after training, ensuring usability. These findings suggest the feasibility and effectiveness of immersive rhythm-based gamified VR systems for cognitive training, laying a foundation for future research on long-term cognitive intervention technologies.

Abstract:
Can force-feedback gloves deliver the haptic fidelity required for medical training or engineering design in VR? We systematically measured stiffness discrimination thresholds using Dexmo force-feedback gloves through psychophysical experiments with 23 participants exploring virtual volumetric data. Our dual-method approach-adaptive staircase procedures and population-level psychometric functions-revealed Just Noticeable Difference (JND) thresholds of 48.1% (Low Reference) and 26.3% (High Reference), far exceeding the 10-20% thresholds achievable with natural touch. This performance gap, compounded by a 1.8:1 Weber fraction ratio indicating perceptual scaling violations, raises questions about current haptic VR capabilities. However, spatial analysis uncovered actionable insights: positioning interactions at mid-reach positions (avoiding near-body and extended-arm extremes) improves accuracy by 7%. Our findings establish critical benchmarks for haptic VR development and provide evidence-based guidelines for designing applications that work within current technological constraints rather than assuming real-world haptic equivalence.

Abstract:
Transcranial magnetic stimulation (TMS) is a noninvasive brain stimulation technique that uses magnetic pulses to safely modulate neural activity in specific brain regions, with particular use in the treatment of major depressive disorder. However, effective treatment protocols require multiple sessions over time and accurate targeting. Current neuronavigation systems can improve pulse delivery compared to standard scalp and measuring tape methods, but are costly and have time-intensive setups for recurring sessions. We present an augmented reality neuronavigation system (AR-NS) that overcomes these limitations. The AR-NS is functionally similar to other neuronavigation systems but can operate entirely within a head-mounted display. Unlike traditional 2D navigation systems that require mentally fusing screen information with real-world actions, the AR interface allows operators to perform the procedure with reduced cognitive and hand-eye coordination demands. We measured the functional targeting accuracy of the AR-NS and the Localite TMS Navigator, a commercial neuronavigation system, using a TMS phantom we developed embedded with Hall effect sensors at four different sites. We determined the coil placement for each site that maximized the sensor response to a magnetic pulse using co-registered X-ray computed tomography and structured light scanner scans. A MagVenture C-B60 coil was placed and fired at each site 30 times in a randomized order according to each neuronavigation system. We measured the resulting magnetic pulse amplitudes. A non-inferiority test with a 5 mT margin and 97.5% confidence intervals indicated that the AR-NS demonstrated similar functional targeting accuracy as the Localite TMS Navigator across all but one stimulation site, indicating that augmented reality neuronavigation systems may offer more accessible delivery of TMS stimulation without sacrificing functional accuracy compared to current systems.

Abstract:
A key factor in successfully achieving remarkable performance for deep learning models is the availability of large amounts of data. However, in scientific visualization, providing such extensive volumetric data is often infeasible due to the high computational cost of simulations and the challenges of data storage. To address this data sparsity issue in model training, we propose a few-shot learning framework that leverages only few training samples (e.g., 1, 3, or 5) to ensure both generalization capability and performance through a conditional diffusion model. Our approach consists of two stages: the forward process and the reverse process. In the forward process, we inject noise at various levels into the few samples. In the reverse process, we design a time-aware UNet that iteratively learns to denoise the noisy data. Additionally, we introduce a noise-aware loss function that dynamically adjusts optimization weights based on the noise levels in the training data. Our method demonstrates consistent and robust performance, regardless of the selection method for the few training samples. Furthermore, it achieves superior results in both quantitative and qualitative evaluations compared to state-of-the-art solutions across three scientific visualization tasks: spatial super-resolution, temporal super-resolution, and variable translation.

Abstract:
Machine Unlearning (MU) aims to remove target training data from a trained model so that the removed data no longer influences the model’s behavior, fulfilling “right to be forgotten” obligations under data privacy laws. Yet, we observe that researchers in this rapidly emerging field face challenges in analyzing and understanding the behavior of different MU methods, especially in terms of three fundamental principles in MU: accuracy, efficiency, and privacy. Consequently, they often rely on aggregate metrics and ad-hoc evaluations, making it difficult to accurately assess the trade-offs between methods. To fill this gap, we introduce a visual analytics system, Unlearning Comparator, designed to facilitate the systematic evaluation of MU methods. Our system supports two important tasks in the evaluation process: model comparison and attack simulation. First, it allows the user to compare the behaviors of two models, such as a model generated by a certain method and a retrained baseline, at class-, instance-, and layer-levels to better understand the changes made after unlearning. Second, our system simulates membership inference attacks (MIAs) to evaluate the privacy of a method, where an attacker attempts to determine whether specific data samples were part of the original training set. We evaluate our system through a case study visually analyzing prominent MU methods and demonstrate that it helps the user not only understand model behaviors but also gain insights that can inform the improvement of MU methods.

Abstract:
Interface metaphors are thought to enable an intuitive and effective interaction with a user interface by allowing users to draw on existing knowledge and reducing the need for instructions. This makes metaphors a promising candidate to drive the development of locomotion interfaces for virtual reality (VR). Since creating metaphoric interfaces can be difficult, it is important to analyze how typical locomotion metaphors can support an intuitive interaction. We performed a qualitative online study to observe the effect of typical metaphors (Walking, Steering, Flying, and Teleportation) and the influence of perceived affordances and user background. Our analysis shows that users adapt the interface expectations induced by metaphors to fit the perceived affordances instead of changing them. Several interests, the age, education, and gender influenced the expectations regarding the VR locomotion interface. Our findings contribute to a better understanding of users’ mental models of VR locomotion metaphors, which seem necessary for designing more intuitive locomotion.

Abstract:
This study explores the design space of two-handed input (i.e., clicking or tapping with the thumb) on the touchpads of controllers for virtual reality (VR) interaction. Four experiments were conducted to fulfill this purpose. Experiment 1 investigated how users employed two VR controllers to perform four representative interaction tasks in VR and identified 14 potentially usable two-handed operations that involved tapping or clicking. Experiments 2 and 3 analyzed user performance of the 14 operations, providing insights into their interaction characteristics in terms of completion time, accuracy, and subjective feedback. In Experiment 4, we designed a command-input technique based on the proposed operations. We verified its effectiveness compared to context menus and marking menus in a VR text entry scenario. Our technique generally had shorter times and similar accuracy to the two menu types. Our work contributes to the design of VR interactions using two-handed controllers.

Abstract:
In applications where efficiency is critical, developers may examine their compiled binaries, seeking to understand how the compiler transformed their source code and what performance implications that transformation may have. This analysis is challenging due to the vast number of disassembled binary instructions and the many-to-many mappings between them and the source code. These problems are exacerbated as source code size increases, giving the compiler more freedom to map and disperse binary instructions across the disassembly space. Interfaces for disassembly typically display instructions as an unstructured listing or sacrifice the order of execution. We design a new visual interface for disassembly code that combines execution order with control flow structure, enabling analysts to both trace through code and identify familiar aspects of the computation. Central to our approach is a novel layout of instructions grouped into basic blocks that displays a looping structure in an intuitive way. We add to this disassembly representation a unique block-based mini-map that leverages our layout and shows context across thousands of disassembly instructions. Finally, we embed our disassembly visualization in a web-based tool, DisViz, which adds dynamic linking with source code across the entire application. DizViz was developed in collaboration with program analysis experts following design study methodology and was validated through evaluation sessions with ten participants from four institutions. Participants successfully completed the evaluation tasks, hypothesized about compiler optimizations, and noted the utility of our new disassembly view. Our evaluation suggests that our new integrated view helps application developers in understanding and navigating disassembly code.

Abstract:
Recent studies have reported a shortcoming of visual exploratory data analysis (EDA) that can lead analysts to report spurious findings. These findings have prompted the advocacy of incorporating belief elicitation within the data analysis process. However, the results from these studies primarily drew from laboratory experiments, which can differ from real-world analysis contexts. In this article, we present outcomes from a longitudinal study with students enrolled in a university-level visual analytics course tackling the VAST Challenge. The students formed teams that were randomly assigned to the belief elicitation and non-belief elicitation conditions. Our study results indicate teams that underwent belief elicitation adopted an intentional approach, while teams in the non-belief elicitation condition reported greater diversity in findings, aligning with prior research. Surprisingly, teams from both conditions achieved equal success in solving the VAST Challenge, suggesting that analysts can incorporate belief elicitation strategically for different goals. We provide guidelines for incorporating belief elicitation into data analysis and teaching material for educators to include belief elicitation in visual analytics courses.

Abstract:
Recovering a continuous colormap from a single 2D scalar field visualization can be quite challenging, especially in the absence of a corresponding color legend. In this paper, we propose a novel colormap recovery approach that extracts the colormap from a color-encoded 2D scalar field visualization by simultaneously predicting the colormap and underlying data using a decoupling-and-reconstruction strategy. Our approach first separates the input visualization into colormap and data using a decoupling module, then reconstructs the visualization with a differentiable color-mapping module. To guide this process, we design a reconstruction loss between the input and reconstructed visualizations, which serves both as a constraint to ensure strong correlation between colormap and data during training, and as a self-supervised optimizer for fine-tuning the predicted colormap of unseen visualizations during inferencing. To ensure smoothness and correct color ordering in the extracted colormap, we introduce a compact colormap representation using cubic B-spline curves and an associated color order loss. We evaluate our method quantitatively and qualitatively on a synthetic dataset and a collection of real-world visualizations from the VIS30K dataset [9]. Additionally, we demonstrate its utility in two prototype applications-colormap adjustment and colormap transfer-and explore its generalization to visualizations with color legends and ones encoded using discrete color palettes.

Abstract:
Modern scientific discovery faces challenges in integrating the rapidly expanding and diverse knowledge required for exploring novel knowledge in biology. While traditional hypothesis-driven research has proven effective, it is constrained by human cognitive limitations, knowledge complexity, and the high costs of trial-and-error experimentation. Deep learning models, particularly graph neural networks (GNNs), have accelerated scientific progress. However, the vast predictions generated make manual selection for experimental validation impractical. Attempts to leverage large language models (LLMs) for filtering predictions and generating novel hypotheses have been impeded by issues such as hallucinations and the lack of structured knowledge grounding, which undermine their reliability. To address these challenges, we propose HypoChainer, a collaborative visualization framework that integrates human expertise, LLM-driven reasoning, and knowledge graphs (KGs) to enhance scientific discovery visually. HypoChainer operates through three key stages: (1) Contextual Exploration: Domain experts employ retrieval-augmented LLMs (RAGs) and visualizations to extract insights and research focuses from vast GNN predictions, supplemented by interactive explanations for in-depth understanding; (2) Hypothesis Construction: Experts iteratively explore the KG information relevant to the predictions and hypothesis-aligned entities, gaining knowledge and insights while refining the hypothesis through suggestions from LLMs; and (3) Validation Selection: Predictions are prioritized based on the refined hypothesis chains and KG-supported evidence, identifying high-priority candidates for validation. The hypothesis chains are further optimized through visual analytics of the retrieval results. We evaluated the effectiveness of HypoChainer in hypothesis construction and scientific discovery through a case study and expert interviews.

Abstract:
Tactile graphics are widely used to present maps and statistical diagrams to blind and low vision (BLV) people, with accessibility guidelines recommending their use for graphics where spatial relationships are important. Their use is expected to grow with the advent of commodity refreshable tactile displays. However, in stark contrast to visual information graphics, we lack a clear understanding of the benefts that well-designed tactile information graphics offer over text descriptions for BLV people. To address this gap, we introduce a framework considering the three components of encoding, perception and cognition to examine the known benefts for visual information graphics and explore their applicability to tactile information graphics. This work establishes a preliminary theoretical foundation for the tactile-frst design of information graphics and identifes future research avenues.

Abstract:
Virtual Reality (VR) broadcasting has emerged as a promising medium for providing immersive viewing experiences of major sports events such as tennis. However, current VR broadcast systems often lack an effective camera language and do not adequately incorporate dynamic, in-game visualizations, limiting viewer engagement and narrative clarity. To address these limitations, we analyze 400 out-of-play segments from eight major tennis broadcasts to develop a tennis-specific design framework that effectively combines cinematic camera movements with embedded visualizations. We further refine our framework by examining 25 cinematic VR animations, comparing their camera techniques with traditional tennis broadcasts to identify key differences and inform adaptations for VR. Based on data extracted from the broadcast videos, we reconstruct a simulated game that captures the players' and ball's motion and trajectories. Leveraging this design framework and processing pipeline, we develope Beyond the Broadcast, a VR tennis viewing system that integrates embedded visualizations with adaptive camera motions to construct a comprehensive and engaging narrative. Our system dynamically overlays tactical information and key match events onto the simulated environment, enhancing viewer comprehension and narrative engagement while ensuring perceptual immersion and viewing comfort. A user study involving tennis viewers demonstrate that our approach outperforms traditional VR broadcasting methods in delivering an immersive, informative viewing experience.

Abstract:
Explorative flow visualization allows domain experts to analyze complex flow structures by interactively investigating flow patterns. However, traditional visual interfaces often rely on specialized graphical representations and interactions, which require additional effort to learn and use. Natural language interaction offers a more intuitive alternative, but teaching machines to recognize diverse scientific concepts and extract corresponding structures from flow data poses a significant challenge. In this paper, we introduce an automated framework that aligns flow pattern representations with the semantic space of large language models (LLMs), eliminating the need for manual labeling. Our approach encodes streamline segments using a denoising autoencoder and maps the generated flow pattern representations to LLM embeddings via a projector layer. This alignment empowers semantic matching between textual embeddings and flow representations through an attention mechanism, enabling the extraction of corresponding flow patterns based on textual descriptions. To enhance accessibility, we develop an interactive interface that allows users to query and visualize flow structures using natural language. Through case studies, we demonstrate the effectiveness of our framework in enabling intuitive and intelligent flow exploration.

Abstract:
Augmented reality (AR) has increasingly been used to communicate environmental impacts, offering greater engagement than conventional displays. However, its effect on message credibility-how much people believe in the content of the communication-remains unclear. In a preregistered study, we compared the perceived credibility of environmental information presented via visualizations on an AR headset or a desktop display. We created display-specific visual encodings (3D concrete for AR, 2D bar charts for desktop) and added two control conditions to cross display and encoding. We found no difference in message credibility between AR and desktop, though concrete AR was rated most engaging. Supplementary material is available at https://osf.io/n4p5c/.

Abstract:
In this work, we challenge the dominant use of logarithmic scales to communicate values spanning multiple orders of magnitude-Orders of Magnitude Values (OMVs)-to the general public. Focusing on bar charts, we incorporate cognitive insights into visualization design to better align with how humans perceive OMVs. Studies in cognitive psychology suggest that, for large numerical ranges such as millions and billions, people do not think logarithmically. Instead, they perceive numbers in a piecewise linear manner, grouping values into scale words (e.g., millions) and applying linear reasoning within each group. We build upon a recently introduced piecewise linear scale, EplusM, and validate its use in bar charts, which we refer to as EplusM bar charts. We also introduce two novel variants of the EplusM bar chart informed by findings in numerical perception: Bricks, which builds on the concepts of round numbers and subitizing, and Multi-Magnitude, which leverages categorical perception of large numbers. In a crowdsourced experiment, we evaluate four bar chart designs: 1) Log, 2) EplusM, 3) Bricks, and 4) Multi-Magnitude, across value retrieval and quantitative comparison tasks. Our results show that EplusM bar charts are significantly preferred over logarithmic designs, increase user confidence, and reduce perceived mental demand, while maintaining task performance. These findings suggest that EplusM bar charts can serve as effective alternatives to logarithmic ones when visualizing OMVs for general audiences.

Abstract:
Connectomics, a subfield of neuroscience, reconstructs structural and functional brain maps at synapse-level resolution. These complex spatial maps consist of tree-like neurons interconnected by synapses. Motif analysis is a widely used method for identifying recurring subgraph patterns in connectomes. These motifs, thus, potentially represent fundamental units of information processing. However, existing computational tools often oversimplify neurons as mere nodes in a graph, disregarding their intricate morphologies. In this paper, we introduce MoMo, a novel interactive visualization framework for analyzing neuron morphology-aware motifs in large connectome graphs. First, we propose an advanced graph data structure that integrates both neuronal morphology and synaptic connectivity. This enables highly efficient, parallel subgraph isomorphism searches, allowing for interactive morphological motif queries. Second, we develop a sketch-based interface that facilitates the intuitive exploration of morphology-based motifs within our new data structure. Users can conduct interactive motif searches on state-of-the-art connectomes and visualize results as interactive 3D renderings. We present a detailed goal and task analysis for motif exploration in connectomes, incorporating neuron morphology. Finally, we evaluate MoMo through case studies with four domain experts, who asses the tool's usefulness and effectiveness in motif exploration, and relevance to real-world neuroscience research. The source code for MoMo is available here.

Abstract:
Pathology images are crucial for cancer diagnosis and treatment. Although artificial intelligence has driven rapid advancements in pathology image analysis, the interpretation of ultra-large and multi-scale pathology images in clinical practice still heavily relies on physicians' experience. Clinicians need to repeatedly zoom in and out on individual slides to compare and assess pathological details - a process that is both time-consuming and prone to visual fatigue. The system first employs a diffusion model to perform tissue segmentation on pathology images, then calculates pathological tissue proportions and morphological metrics. Finally, through multi-scale dynamic comparison and multi-level visual evaluation, the system facilitates comprehensive and precise analysis of pathology images. The system provides clinicians with an intelligent and interactive tool for pathology image interpretation, enabling efficient visualization and precise analysis of pathological details, thereby reducing the effort require for detailed analysis.

Abstract:
Accounting for individual differences can improve the effectiveness of visualization design. While the role of visual attention in visualization interpretation is well recognized, existing work often overlooks how this behavior varies based on visual literacy levels. Based on data from a 235-participant user study covering three visualization tests (mini-VLAT, CALVI, and SGL), we show that distinct attention patterns in visual data exploration can correlate with participants' literacy levels: While experts (high-scorers) generally show a strong attentional focus, novices (low-scorers) focus less and explore more. We then propose two computational models leveraging these insights: Lit2Sal - a novel visual saliency model that predicts observer attention given their visualization literacy level, and Sal2Lit - a model to predict visual literacy from human visual attention data. Our quantitative and qualitative evaluation demonstrates that Lit2Sal outperforms state-of-the-art saliency models with literacy-aware considerations. Sal2Lit predicts literacy with 86% accuracy using a single attention map, providing a time-efficient supplement to literacy assessment that only takes less than a minute. Taken together, our unique approach to consider individual differences in salience models and visual attention in literacy assessments paves the way for new directions in personalized visual data communication to enhance understanding.

Abstract:
Dendroid stony corals build highly complex colonies that develop by asexual reproduction from a single coral polyp sitting in a cup-like exoskeleton, called corallite, resulting in a tree-like branching pattern of its exoskeleton. Despite their beauty and ecological importance as reef builders in tropical shallow-water and in huge cold-water coral mounds in the deep ocean, systematic studies investigating the ontogenetic morphological development of such coral colonies are largely missing. Two main reasons for this lack of study are the large number of corallites, and the existence of many secondary joints/coenosteal bridges in the ideally tree-like structure that make a reconstruction of the skeleton tree extremely tedious. Herein, we present CoDA, the Coral Dendroid structure Analyzer, a visual analytics toolkit that allows for the first time to systematically create skeleton trees representing the correct biological relationship of even very complex dendroid stony corals and to perform ontogenetic morphological analyses based on it. Starting with an initial instance segmentation of the calices/corallites, CoDA estimates the skeleton tree and provides convenient tools and visualizations for proofreading and correcting segmentation and skeleton tree. Part of CoDA is CoDA.Graph, a feature-rich link-and-brush user interface for showing the extracted morphological features and graph layouts of the skeleton tree, enabling real-time exploration of complex coral colonies and their building blocks, the individual corallites and branches. The use of CoDA is exemplified on multiple specimens of the three most important reef-building cold-water coral species with largely varying morphotypes.

Abstract:
In this work, we explore the integration of Orientation Selection into Steering interfaces, aiming to preserve the seamless sensation of real-world movement while mitigating the risk of inducing cybersickness. Our implementation encounters conflicts in standard input mappings, prompting us to adopt bimanual interaction as a solution. Recognizing the complexity of interaction that may arise from this step, we also develop unimanual alternatives, e.g., utilizing a Human-Joystick, commonly referred to as a Leaning interface. The outcomes of an empirical study centered around a primed search task yield unexpected findings. We observed a sample of users spanning multiple levels of gaming experience and a balanced gender distribution exhibit no significant difficulties with the bimanual, asymmetric interfaces. Remarkably, the performance of Orientation Selection is, as in prior work, at least on par with Snap Rotation. Moreover, through a subsequent exploratory analysis, we uncover indications that Pointing-Directed Steering outperforms embodied interfaces in usability and task load in the given setting.

Abstract:
Among recent graph drawing (GD) methods, tsNET creates high quality layouts but suffers from a very high runtime due to its underlying reliance on the t-SNE projection technique. We address this problem by presenting NNP-NET, a method that adapts NNP, a projection technique that can project high-dimensional datasets linearly in the data size, to handle both unweighted and weighted graphs, with layout quality being very close to the ground-truth tsNET. We also exploit NNP’s built-in out-of-sample ability to enable NNP-NET to project time-dependent (dynamic) graphs while striking a good balance between layout stability and good layout quality. We show experiments that outline how NNP-NET can handle very large graphs – up to 50 million nodes and 108 million edges faster than all other comparable methods we are aware of while also yielding good quality metric values.

Abstract:
A truly anytime, anywhere approach to immersive analytics is predicated on the ability for users to author and explore visualizations in an immersive fashion and without a desktop computer. In this paper, we present a generalizable visual programming technique for immersive authoring, exploration, and presentation of data visualizations through direct manipulation, and implement it in a web-based augmented reality (AR) system. In a preregistered user study, we use this implementation to explore how analysts author and use data visualizations in immersive AR accessed using a head-mounted display (HMD). The study focuses on the authoring technique, spatial organization strategies, and sensemaking and presentation practices in such immersive analytics environments for HMD-based AR. Our findings suggest that participants particularly liked the direct manipulation of our approach, and that participants’ spatial organization is not based on physical landmarks in the room.

Abstract:
Radial visualizations encode one dimension of a data set with the visual variable angle or arc length. The radial axis conveys a similarity between the positions of the data points, due to the law of proximity. In the literature, two different types of axes are employed in visualizations: open and closed circular axes. Open axes are linear axes bent to form an open circle. A gap between the right and left poles indicates dissimilarity of data points at both poles. Contrarily, closed axes form a full circle. While the choice of the appropriate axis type in the visualization should align with the similarity space, a question arises: Which axis type supports the human perception of proximity better? To answer this question, we conducted a quantitative task-driven experiment (N=28) to evaluate human distance perception on open and closed circular axes. Within four low-level tasks of three types (identify, compare, and summarize), we evaluate accuracy and response time. Based on our results, we provide an empirically grounded guideline for selecting the appropriate axis type. In our extensive post-hoc analysis, we gain preliminary, but valuable insights to inform further research.

Abstract:
Existing learning-based visual retouching primarily focuses on improving image quality through end-to-end objective mapping between input and retouched images. However, these approaches often overlook two critical aspects: the progressive nature of image retouching and the subjective aesthetic preferences, resulting in suboptimal visual outcomes. To address this, we introduce Automatic Aesthetic Image Retouching via active reinforcement learning (A^33RL) to enhance the visualization experience in two sub-tasks: color enhancement and composition optimization, which are formulated as a unified Markov Decision Process in the proposed A^33RL framework. In our approach, each pixel functions as an autonomous agent that determines optimal actions based on aesthetic guidance, engaging in online exploration through immediate pixel-wise and channel-wise feedback from the aesthetic environment. By leveraging a pretrained image aesthetic model, our method ensures that the A^33RL process aligns with human aesthetic preferences and adheres to subjective aesthetic principles. The framework integrates pixel-level retouching actions with image-level operations to achieve optimal image sequences through progressive iterations. Extensive experiments demonstrate that our method effectively recalibrates image aesthetics across multiple dimensions: low-level quality metrics (PSNR, SSIM), visual perception (LPIPS), and subjective visual experience (human survey). The results demonstrate high consistency with expert-retouched ground-truth images.

Abstract:
Despite much progress, achieving real-time high-fidelity head avatar animation is still difficult and existing methods have to trade-off between speed and quality. 3DMM based methods often fail to model non-facial structures such as eyeglasses and hairstyles, while neural implicit models suffer from deformation inflexibility and rendering inefficiency. Although 3D Gaussian has been demonstrated to possess promising capability for geometry representation and radiance field reconstruction, applying 3D Gaussian in head avatar creation remains a major challenge since it is difficult for 3D Gaussian to model the head shape variations caused by changing poses and expressions. In this paper, we introduce PSAvatar,1 a novel framework for animatable head avatar creation that utilizes discrete geometric primitive to create a parametric shape model and employs 3D Gaussian for fine detail representation and high fidelity rendering. The parametric shape model is a Point-based Shape Model (PSM) which uses points instead of meshes for 3D representation to achieve enhanced representation flexibility. Specifically, PSM first converts the FLAME mesh to points by sampling on the surfaces as well as off the meshes to enable the reconstruction of not only surface-like structures but also complex geometries such as eyeglasses and hairstyles. By aligning these points with the head shape in an analysis-by-synthesis manner, the PSM makes it possible to utilize 3D Gaussian for fine detail representation and appearance modeling, thus enabling the creation of high-fidelity avatars. We show that PSAvatar can reconstruct high-fidelity head avatars of varieties of subjects and the avatars can be animated in real-time.2

Abstract:
Quantum Neural Networks (QNNs) represent a promising fusion of quantum computing and neural network architectures, offering speed-ups and efficient processing of high-dimensional, entangled data. A crucial component of QNNs is the encoder, which maps classical input data into quantum states. However, choosing suitable encoders remains a significant challenge, largely due to the lack of systematic guidance and the trial-and-error nature of current approaches. This process is further impeded by two key challenges: (1) the difficulty in evaluating encoded quantum states prior to training, and (2) the lack of intuitive methods for analyzing an encoder's ability to effectively distinguish data features. To address these issues, we introduce a novel visualization tool, XQAI-Eyes, which enables QNN developers to compare classical data features with their corresponding encoded quantum states and to examine the mixed quantum states across different classes. By bridging classical and quantum perspectives, XQAI-Eyes facilitates a deeper understanding of how encoders influence QNN performance. Evaluations across diverse datasets and encoder designs demonstrate XQAI-Eyes's potential to support the exploration of the relationship between encoder design and QNN effectiveness, offering a holistic and transparent approach to optimizing quantum encoders. Moreover, domain experts used XQAI-Eyes to derive two key practices for quantum encoder selection, grounded in the principles of pattern preservation and feature mapping.

Abstract:
The Simulator Sickness Questionnaire (SSQ) has become a standard tool for quantifying the severity and distribution of discomfort symptoms in virtual reality (VR) research. Despite its straightforward administration, the use of the SSQ also comes with significant challenges, including response subjectivity, strict threshold values based on a military reference population, and confounding factors influencing the results. To demonstrate the adverse interplay of these issues, we asked three cohorts of students to fill in a SSQ after having attended a 90-minute lecture of our teaching program. Although students were not exposed to any form of VR experience, the resulting SSQ scores were indistinguishable from VR studies and extended far beyond the originally defined threshold of a “bad simulator”, with 88.1% of TS scores being larger and 25.4% even exceeding thrice this value. We compare our results to alternative scoring systems of the SSQ proposed in the literature and suggest implications for future experimental designs involving the quantification of sickness symptoms. In summary, our results motivate to exert caution when interpreting the results of the SSQ in the context of a VR study; participants might just have attended a lecture prior to the experiment.

Abstract:
Precise kinematic and dynamic representations are equally critical for fine-grained motor skills such as golf, yet the latter remains underexplored. This work introduces MuscleGolfAR, an augmented reality (AR) training system integrating swing motions and muscle activities. To support cost-effective inference of muscle activations, we construct a multimodal dataset encompassing posture, electromyography (EMG), and plantar pressure. The system displays inferred EMG through embodied (first-person perspective) and detached (third-person perspective) visualizations. User studies are subsequently conducted to evaluate the impact of these strategies on training effectiveness. Results reveal that embodied visualizations enhance ownership over augmented feedback, while detached visualizations facilitate a holistic comprehension of the whole body. Furthermore, individuals' preferences for these strategies correlate with their practice habits and skill proficiencies, offering new insights for the design of AR-based motor skill training systems.

Abstract:
This paper investigates how sinusoidal terrain characteristics and input modalities influence locomotion performance and user experience in virtual reality (VR). Two studies were conducted using a custom-built elastic balance board (EBB). The first study (Study 1) employed a 2 (amplitude: low vs. high) × 2 (frequency: low vs. high) within-group (N = 24) design to examine how slope characteristics affect locomotion efficiency (i.e., speed and completion time), postural stability (i.e., sway velocity), and subjective experiences (i.e., motion sickness, vection, usability, fear of movement, presence, and workload). High frequency slopes in terrains increased completion time, sway velocity, motion sickness, fear of movement, and vection, whereas amplitude alone had minimal effects. The second study (Study 2) used a 2 (amplitude: low vs. high) × 2 (input: EBB vs. joystick) within-group (N = 25) design to explore amplitude effects further and compare locomotion input methods. High amplitude increased completion time, sway velocity, and fear of movement, while EBB use enhanced presence and vection but required a higher workload compared to joystick input. Across both studies, frequency emerged as the primary factor influencing locomotion demands, and amplitude effects became more pronounced when combined with embodied locomotion interfaces. These findings provide design guidelines for VR locomotion systems, emphasizing the importance of aligning terrain complexity with input modality to optimize user comfort and control accuracy.

Abstract:
A prevailing assumption in Virtual Reality is that the perception of surface gloss attenuates with viewing distance due to the resolution limitations of head-mounted displays (HMDs). While prior work indicates that resolution loss can degrade cues such as highlight sharpness and binocular highlight consistency, the direct linkage between viewing distance and perceived gloss has not been systematically established in VR. This study systematically investigates this relationship, focusing on binocular highlight intensity asymmetry. Contrary to the common assumption, under physically based rendering, our results show no reliable decrease in perceived gloss with distance within the tested range, despite resolution limitations. Moreover, participants' gloss judgments appeared to be more strongly modulated by monocular cues (e.g., highlight luminance, contrast, and temporal information) than by the binocular asymmetry manipulation tested here under our tested conditions. Finally, we find that moderate departures from physically correct binocular highlight consistency did not, within a tested tolerance, undermine perceived realism. These findings provide a new basis for designing efficient rendering techniques that preserve gloss fidelity over distance in VR, suggesting that computational resources might be better allocated to preserving monocular cues.

Abstract:
Depth cues, such as binocular disparity, play a critical role in supporting accurate and precise reaching within near space, facilitating reliable interaction with objects and surfaces [27]. Augmented reality (AR) presents unique challenges for such interactions, as mismatched focal distances between real and virtual objects can distort perceived depth. Prior work has examined the perceptual consequences of focal distance mismatches under controlled viewing conditions, as well as AR task performance in more natural settings; here we consider these factors together in a visually guided reaching task with multiple available depth cues for combinations of real and virtual objects. Participants performed a precision visual motor task in an AR environment, placing a virtual ring around either real or virtual posts at varying distances. Our results show that accuracy was degraded when placing a virtual ring on real posts, with fewer successful placements compared to virtual posts. We demonstrate that this increase in error was due to a systematic underestimation of virtual object distances relative to their physical counterparts. These findings indicate that distortions in perceived depth between real and virtual objects persist during active reaching, despite the availability of multiple depth cues including binocular disparity, proprioception, occlusion, and motion parallax. Measuring and correcting for these distortions is important for optimizing the effectiveness and usability of AR devices.

Abstract:
Research on VR often isolates constructs on separate scales, obscuring how users themselves link presence, agency, realism, fictionality, temporal structure, and embodiment during play. We address this gap with a qualitative, participant-level mapping of user-articulated associations across nine constructs defined a priori from theory and standardized for analysis: presence (PI; feeling “there”), visual realism, physical realism, fictionality, agency, temporal structure, narrative comprehension, attentional guidance, and embodiment. Fifty-seven design-trained participants played two commercial VR games (A Fisherman's Tale; Moss on Quest 2/3) and answered openended prompts. Using codebook thematic analysis, we binarized construct mentions per participant and computed normalized co-occurrences (Jaccard) with bootstrap 95% CIs and permutation baselines; metrics are descriptive, valenceagnostic, and non-causal. Findings: Presence co-occurred most with layered attentional guidance and embodiment, exceeding its association with visual fidelity. A”design-side” cluster linked visual and physical realism with temporal pacing and guidance; visual realism clustered more with fictionality and narrative discourse than with presence. These patterns refine theory toward coherence, guidance, and embodiment-centered pathways to sustaining place illusion and engagement, consistent with claims that internal consistency and sensorimotor contingencies outweigh photorealism. Methodologically, the participant-level, normalized map makes qualitative associations inspectable and comparable via effect sizes with uncertainty, complementing solo-construct scales and supporting reuse on other construct sets. Practically, the results recommend prioritizing consistent world rules and physically plausible interactions, scaffolding agency with multimodal diegetic cues, and providing robust, synchronous self-representation.

Abstract:
As immersive Virtual Reality (VR) applications become increasingly widespread, ensuring secure and usable authentication is critical. Traditional knowledge-based methods (e.g., passwords, PINs) suffer from memorability issues and are highly vulnerable to observation attacks such as Man-in-the-Room (MITR). Meanwhile, biometric and behavioral approaches raise concerns regarding practicality, privacy, and cross-platform deployment. We present SemanticAction, a semantic-driven, behavior-augmented authentication framework that integrates knowledge-based passwords with gesture-based behavioral biometrics. Passwords are encoded as scene-anchored directions executed through intuitive hand gestures, enabling semantic meaning to guide user interactions. To counter observation attacks, SemanticAction employs randomized scene prompts and decoy scenes, while a dual-constraint verification mechanism adapts to both cold-start/few-shot conditions. Two user studies provide initial evidence that SemanticAction can mitigate MITR attacks while alleviating memorability challenges, maintaining favorable usability and security performance even with limited behavioral data. This work offers early insights and practical design considerations for behavior-aware authentication in VR.

Abstract:
We propose OFERA, a novel framework for real-time expression control of photorealistic Gaussian head avatars for VR headset users. Existing approaches attempt to recover occluded facial expressions using additional sensors or internal cameras, but sensor-based methods increase device weight and discomfort, while camera-based methods raise privacy concerns and suffer from limited access to raw data. To overcome these limitations, we leverage the blendshape signals provided by commercial VR headsets as expression inputs. Our framework consists of three key components: (1) Blendshape Distribution Alignment (BDA), which applies linear regression to align the headset-provided blendshape distribution to a canonical input space; (2) an Expression Parameter Mapper (EPM) that maps the aligned blendshape signals into an expression parameter space for controlling Gaussian head avatars; and (3) a Mapper-integrated Avatar (MiA) that incorporates EPM into the avatar learning process to ensure distributional consistency. Furthermore, OFERA establishes an end-to-end pipeline that senses and maps expressions, updates Gaussian avatars, and renders them in real-time within VR environments. We show that EPM outperforms existing mapping methods on quantitative metrics, and we demonstrate through a user study that the full OFERA framework enhances expression fidelity while preserving avatar realism. By enabling real-time and photorealistic avatar expression control, OFERA significantly improves telepresence in VR communication. A project page is available at https://ysshwan147.github.io/projects/ofera/.

Abstract:
The Video See-Through (VST) technology in Extended Reality Head-Mounted Displays (XR HMDs) allows users to perceive the real-world environment while viewing seamlessly integrated virtual content in passthrough mode. This technology has demonstrated significant potential across domains such as industry, healthcare, and intelligent driving. However, the camera of current VST technologies often struggles with visual fidelity, especially in low-light environments. To address this challenge, we introduce NoctuaXR, a lightweight, low-cost, environment-aware enhancement that dynamically adjusts the pixel-level brightness and saturation of the VST view in real time based on ambient light conditions, significantly improving visual quality in low-light scenarios. In a 10 lx environment, which is a very dim level of light, our tests show that NoctuaXR improves the color accuracy of the Baseline by 68.59% and outperforms the current leading, professional XR HMD, the Varjo XR-4 Focal Edition, by 64.26%. Furthermore, a comprehensive evaluation, including a perceptual quality evaluation and a daily task performance evaluation, confirms NoctuaXR's effectiveness. Our method provides users with a significantly improved visual experience, elevating their task efficiency to a level comparable to the Varjo XR-4. We also present compelling real-world use cases—from in-flight entertainment to navigating a dark room without disturbing others—that highlight NoctuaXR's tangible value in making XR technology safer, more convenient, and truly practical for daily life.

Abstract:
Optical see-through head-mounted displays (OST-HMDs) allow users to interact with both physical and virtual displays overlaid in the real world within augmented reality (AR) environments. When arranging these displays in space, spatial discontinuity—such as differences in viewing distance or lateral separation in the visual field—may occur. This study examined the effects of displays' spatial discontinuities on multitasking performance and perceived workload in AR, as well as how these effects were influenced by task load. Participants completed a multitasking simulation with four tasks distributed across physical and virtual displays under four display configurations: two configurations with spatial discontinuity, one cross-display configuration without discontinuity, and a baseline configuration where all tasks were shown on the physical display. Each configuration was tested under two task load levels. Results showed that OST-HMD-supported AR significantly impaired multitasking performance, regardless of whether spatial discontinuity between virtual and physical displays was present. Introducing display discontinuities through small depth separations or alignment offsets produced no additional significant influence on performance. The AR configurations also shown increased perceived workload than the baseline, but spatial discontinuity did not significantly affect workload ratings. Task load modulated the effect of display configurations on performance as well as physical and frustration ratings of the perceived workload.

Abstract:
Target selection is a fundamental interaction task in virtual reality (VR) systems, particularly for older adults who face unique challenges due to age-related declines in motor and cognitive abilities. While controller-based raycasting is widely used for its accuracy and efficiency, the design of selection feedback remains an open question, particularly in enhancing usability and accessibility for aging populations. In this study, we propose seven feedback techniques, including three uni-modal (visual, audio, haptic) and four multimodal (visual-audio, visual-haptic, audio-haptic, visual-audio-haptic) approaches. To evaluate these techniques, we conducted two user studies focusing on selection tasks in controlled and realistic scenarios. Our results indicate that visual-based feedback, particularly expansion techniques, significantly improves selection accuracy and user experience. Moreover, multimodal feedback does not always yield better performance; rather, a combination of visual and haptic feedback provides the most effective balance between usability and cognitive load. Based on our findings, we derive six design implications to guide the development of VR selection feedback tailored to older adults. This work contributes to the understanding of optimal selection feedback mechanisms, promoting more inclusive and accessible VR interactions for aging users.

Abstract:
Accessing real-world objects during immersive virtual reality (VR) experiences remains challenging, as current cross-reality systems often rely on predefined interaction steps, tracking devices/markers, or fixed object setups. They also lack support for personalized object recall, where users can add, remove, or modify real-world items blended into the virtual environment (VE). Many head-mounted displays (HMDs) include passthrough technology to switch between virtual and real worlds, but it often disrupts immersion by requiring a full shift from virtual to real. Thus, maintaining an optimal balance between virtuality and reality is difficult. To address these challenges, we developed SelfBlending, a framework that uses AI-based hand tracking to let users label physical objects through freehand gestures, then blends the selected item into the VE using object recognition, enabling interaction with the relevant real-world object. SelfBlending was evaluated against two common interaction conditions: the default passthrough feature in VR HMDs and the conventional approach of physically removing the HMD to access real-world objects. Results from seated, single-object interactions with tabletop-placed items showed that SelfBlending enhanced user experience by boosting presence, supporting efficient physical interaction, and improving cross-reality continuity. It also enabled selective interaction with real objects while minimizing the disruption of VR experience.

Affiliations: HCI Group, University of Konstanz, Konstanz, Germany; Fraunhofer Institute for Process Engineering and Packaging IVV, Dresden, Germany; Interactive Media Lab Dresden, TUD Dresden University of Technology, Dresden, Germany; Aarhus University, Aarhus, Denmark; Columbia University, New York, NY, USA; University of Stuttgart, Stuttgart, Germany; Graz University of Technology, Graz, Austria; Interactive Media Lab Dresden, Centre for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI), Centre for Tactile Internet (CeTI), TUD Dresden University of Technology, Dresden, Germany

Abstract:
We investigate hybrid user interfaces (HUIs), aiming to establish a cohesive understanding and to adopt consistent terminology for this nascent research area. HUIs combine heterogeneous devices in complementary roles, leveraging the distinct benefits of each. Our work focuses on cross-device interaction between 2D devices and mixed reality environments, which are particularly compelling, leveraging the familiarity of traditional 2D platforms while providing spatial awareness and immersion. Although prior work has prominently explored such HUIs in the context of mixed reality, we still lack a cohesive understanding of the unique design possibilities and challenges of such combinations, resulting in a fragmented research landscape. We conducted a systematic survey and present a taxonomy of HUIs that combine conventional display technology and mixed reality environments. Based on this, we discuss past and current challenges, the evolution of definitions, and prospective opportunities to tie together the past 30 years of research with our vision of future HUIs.

Abstract:
Projection Mapping (PM) is a technology that projects images onto the surfaces of physical objects, allowing multiple users to share an augmented reality experience without special devices. However, its practical use has been constrained by the need for dark environments to ensure high-quality projection. To overcome this “dark-room constraint,” we propose a novel target-excluding lighting method that selectively illuminates the surrounding environment while avoiding the PM target. Our system achieves light-field illumination by combining an LED display panel with an optimized aperiodic lens array. The key contributions include a compact form factor that provides a large effective light source area, reproducing natural soft shadows comparable to typical lighting, while maintaining the spatial controllability needed to precisely avoid the target. We also introduce a computational technique for optimizing aperiodic lens placement to suppress undesired dark spots caused by crosstalk, and efficient methods for computing LED luminance patterns that enable dynamic PM. Experiments with a prototype system demonstrate that our approach achieves high-contrast PM even in bright environments.

Abstract:
Aim assistance has been shown to be effective in traditional video games; however, its application in VR contexts remains relatively underexplored. We investigate the effects of invisible bullet-magnetism aim assistance applied at the time of throw on performance and user experience in a VR dart-throwing task. We corrected the trajectory by adjusting the initial throwing direction to reduce the landing position deviation from the target center by 50%. We conducted a within-subjects experiment with 29 participants, comparing the aim-assist condition with a control condition without assistance. Aim assistance with invisible bullet magnetism significantly improved objective and perceived performance, accuracy, and precision. Participants in the aim-assist condition reported significantly higher perceived competence and control, while perceived suspense and pressure were scored significantly lower. Enjoyment, internal attribution of achievement, and workload remained comparable between the two conditions. Participants attributed their success to themselves rather than external factors in both conditions. The invisible bullet-magnetism aim assistance in our study remained largely unnoticed, and participants favored the aim-assist condition.

Abstract:
A fundamental principle of virtual reality is the crafting of illusion which immerses and guides users. In contrast to the ample work on visual illusions and VR, auditory illusions remain relatively understudied. We are particularly interested in how VR can leverage auditory stimuli to manipulate user movement and perception of movement. Here we conduct trials to study how the sounds of footsteps can be manipulated to give the impression of upwards or downwards incline during walking, and how this compares to visual stimuli. We do so by crafting Shepard tone imbued footstep sounds which exploit the cross-modal correspondence between auditory pitch and elevation. We found that while our chosen visual stimuli were considerably more powerful, our footstep sounds still shifted the perception of movement direction, including when they contradicted the direction of visual stimuli. This study opens new research possibilities on how other auditory illusions can influence movement. The Shepard footsteps themselves have direct application in environments where the user must explore the vertical dimension of a 3D space.

Abstract:
Mitigating motion sickness with non-invasive cortical vestibular brain stimulation (vBS) is promising because it does not require redesigning VR content or altering the motion profile of a moving platform. Instead, it aims to enhance individuals' intrinsic resistance to motion sickness. However, the mechanism underlying its online efficacy remains unclear (Here, “online efficacy” means the observed motion sickness reduction during vBS not before or after vBS). Motivated by lower motion sickness susceptibility observed in congenital nystagmus patients, we hypothesized that reductions in motion sickness under vBS condition are associated with increased eye movement instability. Leveraging sham-controlled, sex-balanced eye-tracking data from an open-access database, this study presents the first evaluation of a 20 Hz-vBS approach targeting the vestibular cortices as a countermeasure to motion sickness, using spatiotemporal eye movement metrics derived from eye-tracking data collected in VR. Results show that reduced motion sickness under active vBS indeed increased eye movement instability temporarily and spatially, compared to sham vBS. We also found that the temporal metrics was more strongly associated with female participants, whereas the spatial metric was more strongly associated with male participants.

Abstract:
The use of Augmented Reality (AR) devices for surgical guidance has gained increasing traction in the medical field. Traditional registration methods often rely on external fiducial markers to achieve high accuracy and real-time performance. However, these markers introduce cumbersome calibration procedures and can be challenging to deploy in clinical settings. While commercial solutions have attempted real-time markerless tracking using the native RGB cameras of AR devices, their accuracy remains questionable for medical guidance, primarily due to occlusions and significant outliers between the live sensor data and the preoperative target anatomy point cloud derived from MRI or CT scans. In this work, we present a markerless framework that relies only on the depth sensor of AR devices and consists of two modules: a registration module for high-precision, outlier-robust target anatomy localization, and a tracking module for real-time pose estimation. The registration module integrates depth sensor error correction, a human-in-the-loop region filtering technique, and a robust global alignment with curvature-aware feature sampling, followed by local ICP refinement, for markerless alignment of preoperative models with patient a natomy. The tracking module employs a fast and robust registration algorithm that uses the initial pose from the registration module to estimate the target pose in real-time. We comprehensively evaluated the performance of both modules through simulation and real-world measurements. The results indicate that our markerless system achieves superior performance for registration and comparable performance for tracking to industrial solutions. The two-module design makes our system a one-stop solution for surgical procedures where the target anatomy moves or stays static during surgery.

Abstract:
Room-scale locomotion in VR can promote physical movement to combat the negative effects of a sedentary lifestyle. Therefore, it is useful to investigate factors which may motivate users to walk instead of teleport. In a within-subjects study with 24 participants, we investigated the effects of avatar appearance and target distance on users' locomotion behavior, when given the option to either walk or teleport. In particular, we manipulated three aspects of avatar appearance-body completeness (full-body vs upper-body only), body weight (normal vs obese), and clothing (casual vs superhero armor), and we provided three target distances (1.5 meters, 3.0 meters, and 4.5 meters). We found that avatar obesity decreased willingness to walk, but only for full-body avatars. We also found that close target distances encouraged more walking than teleporting, up to a certain threshold distance. Additionally, we found that users' perceived similarity to their avatar correlated with their willingness to walk. Our findings can be applied to motivate users to walk more in VR, such as in VR exergames.

Abstract:
The emergence of quantum computers heralds a new frontier in computational power, empowering quantum algorithms to address challenges that defy classical computation. However, the design of quantum algorithms is challenging as it largely requires the manual efforts of quantum experts to transit mathematical expressions to quantum circuit diagrams. To ease this process, particularly for prototyping, educational, and modular design workflows, we propose to bridge the textual and visual contexts between mathematics and quantum circuits through visual linking and transitions. We contribute a design space for quantum algorithm design, focusing on the textual and visual elements, interactions, and design patterns throughout the quantum algorithm design process. Informed by the design space, we introduce QuRAFT, a visual interface that facilitates a seamless transition from abstract mathematical expressions to concrete quantum circuits. QuRAFT incorporates a suite of eight integrated visual and interaction designs tailored to support users in the formulation, implementation, and validation process of the quantum algorithm design. Through two detailed case studies and a user evaluation, this paper demonstrates the effectiveness of QuRAFT. Feedback from quantum computing experts highlights the practical utility of QuRAFT in algorithm design and provides valuable implications for future advancements in visualization and interaction design within the quantum computing domain.

Abstract:
Despite the growing availability of 3D urban datasets, extracting insights remains challenging due to computational bottlenecks and the complexity of interacting with data. In fact, the intricate geometry of 3D urban environments results in high degrees of occlusion and requires extensive manual viewpoint adjustments that make large-scale exploration inefficient. To address this, we propose a view-based approach for 3D data exploration, where a vector field encodes views from the environment. To support this approach, we introduce a neural field-based method that constructs an efficient implicit representation of 3D environments. This representation enables both faster direct queries, which consist of the computation of view assessment indices, and inverse queries, which help avoid occlusion and facilitate the search for views that match desired data patterns. Our approach supports key urban analysis tasks such as visibility assessments, solar exposure evaluation, and assessing the visual impact of new developments. We validate our method through quantitative experiments, case studies informed by real-world urban challenges, and feedback from domain experts. Results show its effectiveness in finding desirable viewpoints, analyzing building facade visibility, and evaluating views from outdoor spaces.

Abstract:
The massive data generated by scientists daily serve as both a major catalyst for new discoveries and innovations, as well as a significant roadblock that restricts access to the data. Our paper introduces a new approach to removing Big Data barriers and democratizing access to petascale data for the broader scientific community. Our novel data fabric abstraction layer allows user-friendly querying of scientific information while hiding the complexities of dealing with file systems or cloud services. We enable FAIR (Findable, Accessible, Interoperable, and Reusable) access to datasets such as NASA’s petascale climate datasets. Our paper presents an approach to managing, visualizing, and analyzing petabytes of data within a browser on equipment ranging from the top NASA supercomputer to commodity hardware like a laptop. Our novel data fabric abstraction utilizes state-of-the art progressive compression algorithms and machine-learning insights to power scalable visualization dashboards for petascale data. The result provides users with the ability to identify extreme events or trends dynamically, expanding access to scientific data and further enabling discoveries. We validate our approach by improving the ability of climate scientists to visually explore their data via three fully interactive dashboards. We further validate our approach by deploying the dashboards and simplified training materials in the classroom at a minority-serving institution. These dashboards, released in simplified form to the general public, contribute significantly to a broader push to democratize the access and use of climate data.

Abstract:
Recent advancements in rendering dynamic humans using NeRF and 3D Gaussian splatting have made significant progress, leveraging implicit geometry learning and image appearance rendering to create digital humans. However, in monocular video rendering, there are still challenges in rendering subtle and complex motion from different viewpoints and states, primarily due to the imbalance of viewpoints. Additionally, ensuring continuity between adjacent frames when rendering from novel and free viewpoints remains a difficult task. To address these challenges, we first propose a pixel-level motion correction module that adjusts the errors in the learned representation between different viewpoints. We also introduce a temporal information-based model to improve motion continuity by leveraging adjacent frames. Experimental results on dynamic human rendering, using the NeuMan, ZJU-Mocap, and People-Snapshot datasets, demonstrate that our method outperforms state-of-the-art techniques both quantitatively and qualitatively.

Abstract:
High latency is often introduced due to limited computational capabilities and high hardware demands. It has proven to significantly impair user performance in target selection, a fundamental interaction task. Existing research has established that latency negatively impacts selection times and success rates in 2D interactive systems; however, the underlying behavioral mechanisms remain unclear. This article investigates the effects of latency on selection times, success rates, and endpoint distributions in Virtual Reality (VR) with controller-based raycasting and bare-hand direct touch—the two most common selection methods. Our results from a user study (N = 31) revealed distinct patterns between the two methods, leading to two novel mathematical models that account for latency, target width, and movement amplitude. These two models were validated via a new dataset collected from a second user study (N = 16) and were demonstrated to outperform the existing models. Our findings provide actionable recommendations to mitigate the negative impacts of latency and improve user experience in VR interface design.

Abstract:
Embodied conversational agents (ECAs) are key social interaction partners in various virtual reality (VR) applications, with their perceived social presence significantly influencing the quality and effectiveness of user-ECA interactions. This article investigates the potential of a novel indirect objective proxy for evaluating social presence, which is traditionally assessed through subjective questionnaires. As proxy we use the Heard Text Recall (HTR) paradigm, originally designed to assess memory performance in listening tasks. In the task participants have to recall information from family stories being told. When combined with a secondary task in a dual-task paradigm, the HTR can also be used to evaluate cognitive spare capacity, which is then correlated with subjectively rated social presence. As a prerequisite for this investigation, we introduce various co-verbal gesture modification techniques and assess their impact on the perceived naturalness of the presenting ECA, a crucial aspect fostering social presence. The main study then explores the applicability of HTR as a proxy for social presence by examining its effectiveness under different multimodal degraders of ECA behavior, including degraded co-verbal gestures, omitted lip synchronization, and the use of synthetic voices. The findings suggest that while HTR shows potential as an objective measure of social presence, its effectiveness is primarily evident in response to substantial changes in ECA behavior. Additionally, the study also highlights the negative effects of synthetic voices and inadequate lip synchronization on perceived social presence, emphasizing the need for careful consideration of these elements in ECA design.

Abstract:
Affective visualization design is an emerging research direction focused on communicating and influencing emotion through visualization. However, as revealed by previous research, this area is highly interdisciplinary and involves theories and practices from diverse fields and disciplines, thus awaiting analysis from more fine-grained angles. To address this need, this work focuses on a pioneering and relatively mature sub-area, affective geovisualization design, to further the research in this direction and provide more domain-specific insights. Through an analysis of a curated corpus of affective geovisualization designs using the Person-Process-Place (PPP) model from geographic theory, we derived a design taxonomy that characterizes a variety of methods for eliciting and enhancing emotions through geographic visualization. We also identified four underlying high-level design paradigms of affective geovisualization design (e.g., computational, anthropomorphic) that guide distinct approaches to linking geographic information with human experience. By extending existing affective visualization design frameworks with geographic specificity, we provide additional design examples, domain-specific analyses, and insights to guide future research and practices in this underexplored yet highly innovative domain.

Abstract:
London's famous “tube map” is an iconic piece of design and perhaps represents the schematic visualization style most well-known to the general public: its octolinearity has become the de facto standard for transit maps around the world. Making a good schematic transit map is challenging and labour-intensive, and has attracted the attention of the optimization community. Much of the literature has focused on mathematically defining an optimal drawing and algorithms to compute one. However, achieving these “optimal” layouts is computationally challenging, often requiring multiple minutes of runtime. Crucially, what it means for a map to be good is actually highly dependent on factors that evade a general formal definition, like unique landmarks within the network, the context in which a map will be displayed, and the preference of the designer and client. Rather than attempting to make an algorithm that produces a single high-quality and ready-to-use metro map at great cost, we propose it is more fruitful to support rapid layout iteration by a human designer, providing a workflow that enables efficient exploration of a wider range of designs than could be done by hand, and iterating on these designs. To this end we identify steps in the design process of schematic maps that are tedious to do by hand but are algorithmically feasible, and present a framework around a simple linear program that computes network layouts almost instantaneously given a fixed direction for every connection. These connection directions are decided by a designer in a graphical user interface with several interaction methods and a number of quality-of-life features demonstrating the flexibility of the framework; the implementation is available as open source.

Abstract:
Sonification offers a non-visual way to understand data, with pitch-based encodings being the most common. Yet, how well people perceive slope and acceleration—key features of data trends—remains poorly understood. Drawing on people's natural abilities to perceive tempo, we introduce a novel sampling method for pitch-based sonification to enhance the perception of slope and acceleration in univariate functions. While traditional sonification methods often sample data at uniform x-spacing, yielding notes played at a fixed tempo with variable pitch intervals (Variable Pitch Interval), our approach samples at uniform y-spacing, producing notes with consistent pitch intervals but variable tempo (Variable Tempo). We conducted psychoacoustic experiments to understand slope and acceleration perception across three sampling methods: Variable Pitch Interval, Variable Tempo, and a Continuous (no sampling) baseline. In slope comparison tasks, Variable Tempo was more accurate than the other methods when modulated by the magnitude ratio between slopes. For acceleration perception, just-noticeable differences under Variable Tempo were over 13 times finer than with other methods. Participants also commonly reported higher confidence, lower mental effort, and a stronger preference for Variable Tempo compared to other methods. This work contributes models of slope and acceleration perception across pitch-based sonification techniques, introduces Variable Tempo as a novel and preferred sampling method, and provides promising initial evidence that leveraging timing can lead to more sensitive, accurate, and precise interpretation of derivative-based data features.

Abstract:
Examining vision-language alignment in multimodal embeddings is crucial for various tasks, such as evaluating generative models and filtering pretraining data. The intricate nature of high-dimensional features necessitates dimensionality reduction (DR) methods to explore alignment of multimodal embeddings. However, existing DR methods fail to account for cross-modal alignment metrics, resulting in severe occlusion of points with divergent metrics clustered together, inaccurate contour maps from over-aggregation, and insufficient support for multi-scale exploration. To address these problems, this paper introduces DKMap, a novel DR visualization technique for interactive exploration of multimodal embeddings through Dynamic Kernel enhanced projection. First, rather than performing dimensionality reduction and contour estimation sequentially, we introduce a kernel regression supervised t-SNE that directly integrates post-projection contour mapping into the projection learning process, ensuring cross-modal alignment mapping accuracy. Second, to enable multi-scale exploration with dynamic zooming and progressively enhanced local detail, we integrate validation-constrained a refinement of a generalized t-kernel with quad-tree-based multi-resolution technique, ensuring reliable kernel parameter tuning without overfitting. DKMap is implemented as a multi-platform visualization tool, featuring a web-based system for interactive exploration and a Python package for computational notebook analysis. Quantitative comparisons with baseline DR techniques demonstrate DKMap's superiority in accurately mapping cross-modal alignment metrics. We further demonstrate generalizability and scalability of DKMap with three usage scenarios, including visualizing million-scale text-to-image corpus, comparatively evaluating generative models, and exploring a billion-scale pretraining dataset.

Abstract:
Storyline visualizations represent character interactions over time. When these characters belong to different groups, a new research question emerges: how can we balance optimization of readability across the groups while preserving the overall narrative structure of the story? Traditional algorithms that optimize global readability metrics (like minimizing crossings) can introduce quality biases between the different groups based on their cardinality and other aspects of the data. Visual consequences of these biases are: making characters of minority groups disproportionately harder to follow, and visually deprioritizing important characters when their curves become entangled with numerous secondary characters. We present F2Stories, a modular framework that addresses these challenges in storylines by offering three complementary optimization modes: (1) fairnessMode ensures that no group bears a disproportionate burden of visualization complexity regardless of their representation in the story; (2) focusMode allows prioritizing a group of characters while maintaining good readability for secondary characters; and (3) standardMode globally optimizes classical aesthetic metrics. Our approach is based on Mixed Integer Linear Programming (MILP), offering optimality guarantees, precise balancing of competing metrics through weighted objectives, and the flexibility to incorporate complex fairness concepts as additional constraints without the need to redesign the entire algorithm. We conducted an extensive experimental analysis to demonstrate how F2Stories enables more fair or focus group-prioritized storyline visualizations while maintaining adherence to established layout constraints. Our evaluation includes comprehensive results from a detailed case study that shows the effectiveness of our approach in real-world narrative contexts. An open access copy of this paper and all supplemental materials are available at osf.io/e2qvy.

Abstract:
Understanding how people perceive visualizations is crucial for designing effective visual data representations; however, many heuristic design guidelines are derived from specific tasks or visualization types, without considering the constraints or conditions under which those guidelines hold. In this work, we aimed to assess existing design heuristics for categorical visualization using well-established psychological knowledge. Specifically, we examine the impact of the subitizing phenomenon in cognitive psychology-people's ability to automatically recognize a small set of objects instantly without counting-in data visualizations. We conducted three experiments with multi-class scatterplots-between 2 and 15 classes with varying design choices-across three different tasks-class estimation, correlation comparison, and clustering judgments-to understand how performance changes as the number of classes (and therefore set size) increases. Our results indicate if the category number is smaller than six, people tend to perform well at all tasks, providing empirical evidence of subitizing in visualization. When category numbers increased, performance fell, with the magnitude of the performance change depending on task and encoding. Our study bridges the gap between heuristic guidelines and empirical evidence by applying well-established psychological theories, suggesting future opportunities for using psychological theories and constructs to characterize visualization perception.

Abstract:
Light field (LF) cameras encode dense spatial-angular information for depth estimation, critical for applications such as 3D reconstruction, refocusing, and virtual reality. However, current deep learning methods for LF depth estimation still face significant challenges due to the discretization gap between the continuous geometry of real scenes and the discrete sampling of digital images, limiting their effectiveness in high-precision application. This gap manifests in two complementary forms: spatial discretization leads to structural ambiguities and information loss, while depth discretization introduces inaccuracies due to fixed, discrete depth sampling. To address these challenges, we propose Unified Continuous Geometry Representation (UCGR), a unified representation that models scene geometry as a continuous field over image coordinates and depth. UCGR treats spatial and depth discretization as two facets of the same problem and realized by two complementary operators: (1) Adaptive Plane Sampling Operator, which learns edge-aware planar priors to preserve geometric details and mitigate spatial discretization. (2) Contextual Depth Correction Operator, which utilizes contextual information for depth correction, ensuring continuous depth estimation and suppressing artifacts. Building on UCGR, we propose a Continuous Geometry Network (CGNet) that collaboratively optimizes both spatial and depth discretization for accurate and consistent LF depth estimation. Extensive experiments on synthetic and real-world LF datasets demonstrate that CGNet achieves state-of-the-art performance, significantly outperforming existing LF depth estimation methods in terms of both accuracy and robustness.

Abstract:
Driven by advances in supercomputing, the scale of scientific simulation data has grown dramatically. In fields such as cosmology, particle data have become a common representation, with state-of-the-art simulations now exceeding the trillion-particle mark. Consequently, the challenge of visually analyzing such massive datasets has become increasingly urgent. The traditional visual analysis workflow typically follows a “compression \rightarrow→ storage \rightarrow→ reconstruction \rightarrow→ visualization” pipeline. However, this process is hampered by an extremely time-consuming reconstruction stage, which severely impedes real-time interactive visualization. Moreover, in multi-time-step analyses, the enormous volume of reconstructed data creates significant I/O bottlenecks. In this work, we draw inspiration from 3D Gaussian splatting and compress the simulation data using Gaussian Mixture Models (GMMs), treating the resulting Gaussian kernels as fundamental rendering primitives. Our method renders billion-scale particles for each timestep in approximately 32 ms, requiring only 645 MB of GPU memory per timestep — nearly 20× smaller than the original 12 GB raw data. This eliminates costly reconstruction, accelerates the visual analysis pipeline, and overcomes I/O bottlenecks in multi-time-step analysis. Extensive experiments and comparisons across multiple datasets validate the effectiveness of our method.

Abstract:
Early mobilization is a structured protocol designed to facilitate motor recovery in intensive care unit (ICU) patients with ICU-acquired weakness. This process is typically implemented by an interdisciplinary team of nurses, physical therapists, and other healthcare professionals. However, its application is often constrained by the patients' critical conditions, limited mobility, and the challenges of coordinating care within resource-intensive ICU environments. In this study, we developed a patient-centered virtual reality (VR) exergame through an interdisciplinary design process involving clinicians and therapists, tailored to the constraints of critical care. The exergame incorporates progressive mobility levels that mirror early mobilization practices, and includes an embodied avatar to provide guidance and motivation. Using Meta Quest 3 body tracking, the system captures and visualizes patients' movements, thereby providing motivational engagement and quantifiable mobility metrics. We evaluated the exergame in two stages: a dual-user study involving healthy participants and healthcare professionals or students (N = 13), and a subsequent study with cardiothoracic ICU patients (N = 18) to assess feasibility, design validity, and clinical acceptance. Across both studies, participants reported high enjoyment and engagement without discomfort or stress. Furthermore, patients demonstrated increases in movement speed, range of motion, and workspace volume of the upper body across game levels. Physiological monitoring further indicated that the exergame elicited exertion without inducing excessive cardiovascular responses. These findings highlight the feasibility of VR exergames as a clinically acceptable and engaging adjunct to early mobilization in critical care, offering a novel pathway to improve rehabilitation outcomes for ICU patients.

Abstract:
Social Virtual Reality (SVR) allows users to meet and build relationships through embodied avatars and real-time interaction in virtual spaces. While embodiment can strengthen social connections and presence, it can also intensify negative encounters, making SVR particularly vulnerable to harassment. Despite frequent reports of verbal, visual, and “physical” violations in SVR, little is known about how harassment reshapes users' self-perception, including their sense of embodiment, self-identification, closeness, and avatar customization preferences. We conducted a controlled experiment with 52 participants who experienced either a neutral or a harassment condition in a scenario modeled after real SVR incidents. Participants perceived the harassing peer as significantly more negative, annoying, and disturbing than the neutral peer. Contrary to prior reports, harassment did not significantly affect well-being measures, including emotional state, self-esteem, and physiological arousal, within this controlled scenario. However, participants reported stronger bodily change, attributed more of their own attitudes and emotions to their avatars, and increased interpersonal distance when personal space was invaded. Self-reported coping strategies included ignoring, stepping back, using humor, and retaliating. Notably, avatar customization preferences shifted across conditions. Participants in the neutral condition favored personalized avatars, whereas those in the harassment condition more frequently preferred anonymity in public spaces. Together, these findings demonstrate that harassment in SVR not only exploits embodiment but also reshapes self-perception. We further contribute methodological insights into how harassment can be ethically and reproducibly studied in controlled SVR-like experiments.

Abstract:
The adoption of immersive virtual reality (IVR) for gamified rehabilitation is increasing. However, a significant challenge relates to perceptions of virtual environments as empty and isolating, potentially increasing stress, particularly among older users (patients). This paper explores the use of a virtual peer mentor (VPM) in a custom IVR-rehabilitation game to provide guidance and companionship. This game specifically targets patients recovering from acute myocardial infarction (AMI), commonly known as a heart attack. Grounded in social support theory, the VPM provides three types of support: (1) informational support, through pre-exercise narratives detailing a shared medical history; (2) instrumental support, through real-time demonstrations of clinically-validated exercise movements; and (3) emotional support, through positive feedback and encouragement. A within-subjects study involving 30 hospitalized AMI patients (all over 47 years old) evaluated the effectiveness of the VPM-integrated IVR-rehabilitation game. Each participant experienced a baseline (no VPM) and VPM-integrated version of the game on separate days. The results from the Intrinsic Motivation Inventory (IMI) and the social presence module of the Game Experience Questionnaire (GEQ-SPM) show that the VPM resulted in significant increases in engagement, and statistically significant lower pressure/tension. Furthermore, participants exhibited high user acceptance (76.7%) and task-completion rates (98.5%), with minimal cybersickness. The findings demonstrate that a psychologically-grounded VPM can effectively reduce stress in middle- and older-age patients in an IVR rehabilitation setting.

Abstract:
With recent advancements in augmented reality (AR) display technologies, various head-mounted AR displays are now on the consumer market (e.g., Microsoft HoloLens 2, Meta Quest 3). However, there is limited knowledge on how these modern displays impact depth perception. Moreover, understanding the significance of occlusion's effects on action-space distance perception in AR is important, particularly for practical applications such as AR-assisted search-and-rescue operations. However, to our knowledge, no prior research has examined how varying occluder surfaces affect action-space distance estimation across head-mounted AR devices (optical see-through (OST) and video see-through (VST)). In this paper, we present the first research to examine how different occluding surfaces affect action-space distance estimation using OST and VST head-mounted AR devices. With 28 participants across two experiments, we examined verbal estimates and blind walking distance estimation methods. Three occluding surfaces: a checkerboard (high-salient), a black surface (low-salient), and a realistic brick wall texture were considered. Five action-space distances, ranging from 1m to 5m, were evaluated using two commercial AR devices: Microsoft HoloLens 2 (OST) and Meta Quest 3 (VST). Our findings suggest that occluders do not impact AR object action-space distance estimations through blind walking and verbal distance estimation tasks with two modern AR devices. Blind walking distance estimations varied between the HoloLens 2 and Meta Quest 3, with the HoloLens 2 (OST) achieving higher accuracy (distance error: 0.4cm to 3m), than the Meta Quest 3 (VST) (distance error: 24cm to 3m), regardless of occlusion. For AR objects placed on or behind an occluder, distance estimations were overestimated with Meta Quest 3, while the HoloLens 2 either achieved close to accurate or underestimated of depth perception via blind walking.

Abstract:
Cooling feedback has been shown to improve user experience in virtual reality. However, many existing methods face limitations, such as requiring skin contact, relying on airflow, having bulky components, or delivering unstable sensations. These issues limit their applicability to wearable systems, especially in scenarios demanding high responsiveness and perceptual consistency. We propose ChillFactor, a highly responsive, non-contact cooling interface for head-mounted displays (HMDs). It uses ultrasonic atomization of a predetermined alcohol solution to generate evaporative cooling near the skin, reducing airflow-induced artifacts and providing relatively consistent cold sensations. To validate the design, we first evaluated the system's cooling performance. At maximum output, it lowered skin temperature by approximately 2.6 °C within five seconds, a rate sufficient to induce an almost immediate perception of cold. We subsequently integrated ChillFactor into an HMD and conducted a user study with 12 participants in immersive VR scenarios. Participants reported perceiving the cold within 0.66 seconds and noted significantly higher levels of presence and realism compared to conditions without cooling feedback, demonstrating the system's effectiveness in enhancing immersion through rapid and adjustable thermal cues. This work provides a practical foundation for non-contact cooling feedback and expands the design space for multisensory interaction in virtual environments.

Abstract:
Haptic interaction is crucial for immersive virtual physics experiments, but current haptic devices cannot provide high-fidelity feedback by simultaneously simulating realistic physical shapes and dynamic forces. To address this, we introduce PhysicsHap, a modular haptic framework designed for the rapid construction of reconfigurable proxies that provide both high-fidelity shape replication and dynamic force feedback for immersive virtual physics experiments. The framework's core comprises passive Tangible Bodies to simulate physical shape and active Force Engines to deliver dynamic forces. Furthermore, we leverage the hand-tracking technology of head-mounted displays to achieve precise pose synchronization between the physical proxy and its digital twin. We conducted a randomized controlled experiment with 64 secondary school students to evaluate the effectiveness of PhysicsHap in a coupled oscillators system experiment. The results demonstrate that, compared to a non-haptic gesture-based interaction, the PhysicsHap proxies significantly enhanced students' acquisition of experimental physics knowledge and comprehensive abilities, increased their learning motivation, and optimized their cognitive load by reducing extraneous and increasing germane cognitive load. We open-source PhysicsHap to the community at https://github.com/lyh-12/PhysicsHap.

Abstract:
Mixed reality (MR) notification systems currently display all messages in fixed central locations regardless of urgency, leading to unnecessary interruptions and cognitive overload. Drawing from previous MR/Virtual Reality (VR) notification design work and calm technology principles, we developed an adaptive notification system that adjusts spatial placement based on urgency levels: non-urgent notifications appear as peripheral icons accessible via head movement, moderately urgent messages anchor to the user's hand, and very urgent notifications transition progressively from peripheral to central view. Through a within-subjects study (N=18), we evaluated our adaptive system against the default centralised approach. Results demonstrate that the adaptive system significantly reduces mental workload (p=0.041), temporal workload (p=0.008), and frustration (p=0.004) while maintaining comparable notification awareness. Logistic regression analysis reveals that users prefer the adaptive system even with classification errors, provided the combined misclassification rate (disruptiveness + omission errors) remains below a determinable threshold. Our findings establish the first empirical evidence that urgency-based spatial notification distribution effectively addresses core MR usability challenges, offering practical design guidelines for immersive notification systems that balance user attention management with information accessibility.

Abstract:
The perceived realism of virtual body illusions, such as arm elongation, is dependent on congruent sensory feedback. This paper explores the effects of visio-haptic stimulation on the sense of embodiment in Virtual Reality (VR). We conducted three experiments. The first identified the optimal parameters for vibrotactile feedback by varying stimulation location and motor activation duration (250, 500, 750 ms) during a fixed visual task of 40% arm elongation over 18 seconds. We found that sequential stimulations along the entire arm with a 500ms duration per motor were the most effective. From this, we derived a formula to generalize the optimal haptic timing based on elongation velocity. The formula was then validated in two follow-up experiments: one varying the elongation percentage (26% to 80%) and another varying the elongation duration (9s to 27s). Our findings offer validated guidelines for designing more immersive haptic experiences for virtual body deformations.

Abstract:
Historically, distances are underestimated in virtual environments, and a mismatch between a participant's inter-pupillary distance (IPD) and the inter-axial distance (IAD) of the head-mounted display (HMD) could contribute to underestimation, but prior work has not examined how this interacts with lens type or target location in the visual field. Prior work testing the effects of IPD mismatch on distance perception in immersive virtual environments (IVEs) shows mixed results for distances within 2–10 meters. One limitation is that this work did not consider the variety of optical lenses employed in different HMDs. The lens type could affect distance perception differently depending on IPD and could affect it even more for targets appearing on the sides of the display rather than in front of the user. We examined whether the HMD type and location of targets in the virtual visual field affect perception when IPD is mismatched and matched. Using a blind walking task, participants judged the distance to targets that were either in front or to the side of the participant. One group viewed targets in the Varjo XR-3 and the other viewed them in the HTC Vive Pro. We found that distances were judged to be farther in the periphery than in front of users for both devices. Only in the HTC Vive Pro did an increase in IPD mismatch result in increased distance estimates for targets presented directly in front of the participant. Directional angular error (error observed in orienting to the target) increased as the target angle increased for both devices. However, in the Varjo XR-3, a higher IPD mismatch resulted in greater angular error when viewing targets to the side. Overall, the findings show that a mismatched IPD can affect depth and orientation judgments depending on target position in the display. However, there are differences in both FOV and resolution that could influence distance and orientation perception. A second experiment was conducted to examine whether these factors contributed to perception. To test for this, we down-sampled the Varjo XR-3 to match the horizontal field of view and resolution of the HTC Vive Pro. We found that orientation error was mostly affected by resolution and FOV as the adjusted Varjo XR-3 yielded similar orientation perception as the HTC Vive Pro. With this knowledge, researchers should consider HMDs, IPD, and viewing area in the display as possible influences on distance and orientation perception in order to make VR accessible and accurate to all users.

Abstract:
Although haptic technologies that allow users to touch virtual objects have been actively researched in the academic field, they are not yet widely adopted in real-world applications. To promote broader adoption, designing haptic stimuli that elicit pleasant sensations is essential since hedonic motivation significantly drives user acceptance. However, how such pleasantness can be effectively induced through haptic stimulation during a user's active exploration remains an open question. In this study, we investigated the relationship between the applied force during user actions and the resulting pleasantness, using a bow-dragging and arrow-shooting scenario. Specifically, we conducted three psychophysical experiments to investigate two types of force feedback. One type was a physical force that stimulated the kinesthetic sense, and the other was a pseudo force induced through visual manipulation. Our results revealed that pleasantness tended to be elicited when physical resistance force was present during bow dragging, whereas this tendency was not observed for pseudo resistance force. Our findings show that the two types of force are not interchangeable for eliciting positive affect. We also found that the arrow's flying speed contributed to pleasantness as much as the physical resistance force, suggesting that pleasantness is influenced postdictively. Moreover, across all experiments, perceived pleasantness was strongly correlated with the perceived naturalness of the bow-and-arrow behavior. These findings provide valuable insights for designing pleasant interactions involving continuous force feedback, highlighting the distinct roles of physical and pseudo-haptic cues.

Abstract:
We present a novel Mixed Reality (MR) remote collaboration system that integrates a telepresence drone to overcome viewpoint limitations in previous approaches. Current remote collaboration systems often restrict remote users to the local user’s perspective or a fixed viewpoint, which may reduce spatial awareness and collaboration effectiveness. Our proposed system enables a remote user wearing a VR headset to independently navigate the local environment through a drone while guiding a local AR headset user. The drone represents the remote user physically and virtually with an AR avatar. The system provides the remote user with a 3D reconstruction of the local environment, the local user’s avatar, and real-time drone camera video to facilitate drone operation and collaboration. A user study comparing our drone-enhanced MR system with 2D video and 360-degree video MR systems during inspection tasks revealed that the telepresence drone significantly improved collaboration efficiency and enhanced users’ social presence and spatial presence. However, the remote users perceived a higher workload. Our findings demonstrate the benefits of independent spatial navigation through drones in remote collaboration and offer insights for future drone-enhanced MR remote collaboration systems.

Abstract:
Visualizing a large-scale volumetric dataset with high resolution is challenging due to the substantial computational time and space complexity. Recent deep learning-based image inpainting methods significantly improve rendering latency by reconstructing a high-resolution image for visualization in constant time on GPU from a partially rendered image where only a portion of pixels go through the expensive rendering pipeline. However, existing solutions need to render every pixel of either a predefined regular sampling pattern or an irregular sample pattern predicted from a low-resolution image rendering. Both methods require a significant amount of expensive pixel-level rendering. In this work, we provide Importance Mask Learning (IML) and Synthesis (IMS) networks, which are the first attempts to directly synthesize important regions of the regular sampling pattern from the user’s view parameters, to further minimize the number of pixels to render by jointly considering the dataset, user behavior, and the downstream reconstruction neural network. Our solution is a unified framework to handle various types of inpainting methods through the proposed differentiable compaction/decompaction layers. Experiments show our method can further improve the overall rendering latency of state-of-the-art volume visualization methods using reconstruction neural network for free when rendering scientific volumetric datasets. Our method can also directly optimize the off-the-shelf pre-trained reconstruction neural networks without elongated retraining.

Abstract:
Traditional volume visualization (VolVis) methods, like direct volume rendering, suffer from rigid transfer function designs and high computational costs. Although novel view synthesis approaches enhance rendering efficiency, they require additional learning effort for non-experts and lack support for semantic-level interaction. To bridge this gap, we propose NLI4VolVis, an interactive system that enables users to explore, query, and edit volumetric scenes using natural language. NLI4VolVis integrates multi-view semantic segmentation and vision-language models to extract and understand semantic components in a scene. We introduce a multi-agent large language model architecture equipped with extensive function-calling tools to interpret user intents and execute visualization tasks. The agents leverage external tools and declarative VolVis commands to interact with the VolVis engine powered by 3D editable Gaussians, enabling open-vocabulary object querying, real-time scene editing, best-view selection, and 2D stylization. We validate our system through case studies and a user study, highlighting its improved accessibility and usability in volumetric data exploration. We strongly recommend readers check out our case studies, demo video, and source code at https://nli4volvis.github.io/.

Abstract:
Misleading visualizations pose a significant challenge to accurate data interpretation. While recent research has explored the use of Large Language Models (LLMs) for detecting such misinformation, practical tools that also support explanation and correction remain limited. We present MisVisFix, an interactive dashboard that leverages both Claude and GPT models to support the full workflow of detecting, explaining, and correcting misleading visualizations. MisVisFix correctly identifies 96% of visualization issues and addresses all 74 known visualization misinformation types, classifying them as major, minor, or potential concerns. It provides detailed explanations, actionable suggestions, and automatically generates corrected charts. An interactive chat interface allows users to ask about specific chart elements or request modifications. The dashboard adapts to newly emerging misinformation strategies through targeted user interactions. User studies with visualization experts and developers of fact-checking tools show that MisVisFix accurately identifies issues and offers useful suggestions for improvement. By transforming LLM-based detection into an accessible, interactive platform, MisVisFix advances visualization literacy and supports more trustworthy data communication.

Abstract:
Color is a powerful tool in data visualization, but for individuals with color vision deficiencies (CVD), hue can become a barrier rather than an aid. In this paper, we examine how real-world visualizations are perceived across vision profiles through three complementary studies. Study 1 assessed how normal vision participants rated 46 visualizations shown in original and simulated red/green colorblind versions. Study 2 collected matched responses from participants with diagnosed CVD. Study 3 involved in-depth interviews exploring how users interpret, adapt to, and evaluate inaccessible designs. Across studies, we find that simulations capture directional perceptual shifts but fail to reflect the interpretive breakdowns and emotional work described by real CVD users. Factor analysis reveals two dominant perceptual dimensions: functional utility and affective experience. While normal vision participants prioritize functional clarity, CVD users rely more on structural cues and emotional resonance, particularly when color is unreliable. Qualitative insights show that perceptual breakdowns occur not only in high-interference charts but also when redundant encoding or layout scaffolding is missing. We synthesize these findings and offer empirically grounded design recommendations to guide inclusive visualization practices. Our results argue that accessibility must go beyond color correction, embracing structural clarity, redundancy, and real-user validation to ensure inclusive visual communication.

Abstract:
Visualizations support critical decision making in domains like health risk communication. This is particularly important for those at higher health risks and their care providers, allowing for better risk interpretation which may lead to more informed decisions. However, the kinds of visualizations used to represent data may impart biases that influence data interpretation and decision making. Both continuous representations using bar charts and discrete representations using icon arrays are pervasive in health risk communication, but express the same quantities using fundamentally different visual paradigms. We conducted a series of studies to investigate how bar charts, icon arrays, and their layout (juxtaposed, explicit encoding, explicit encoding plus juxtaposition) affect the perception of value comparison and subsequent decision-making in health risk communication. Our results suggest that icon arrays and explicit encoding combined with juxtaposition can optimize for both accurate difference estimation and perceptual biases in decision making. We also found misalignment between estimation accuracy and decision making, as well as between low and high literacy groups, emphasizing the importance of tailoring visualization approaches to specific audiences and evaluating visualizations beyond perceptual accuracy alone. This research contributes empirically-grounded design recommendations to improve comparison in health risk communication and support more informed decision-making across domains.

Abstract:
Multi-agent workflows have become an effective strategy for tackling complicated tasks by decomposing them into multiple sub-tasks and assigning them to specialized agents. However, designing optimal workflows remains challenging due to the vast and intricate design space. Current practices rely heavily on the intuition and expertise of practitioners, often resulting in design fixation or an unstructured, time-consuming exploration of trial-and-error. To address these challenges, this work introduces FLOWFORGE, an interactive visualization tool to facilitate the creation of multi-agent workflow through i) a structured visual exploration of the design space and ii) in-situ guidance informed by established design patterns. Based on formative studies and literature review, FLOWFORGE organizes the workflow design process into three hierarchical levels (i.e., task planning, agent assignment, and agent optimization), ranging from abstract to concrete. This structured visual exploration enables users to seamlessly move from high-level planning to detailed design decisions and implementations, while comparing alternative solutions across multiple performance metrics. Additionally, drawing from established workflow design patterns, FLOWFORGE provides context-aware, in-situ suggestions at each level as users navigate the design space, enhancing the workflow creation process with practical guidance. Use cases and user studies demonstrate the usability and effectiveness of FLOWFORGE, while also yielding valuable insights into how practitioners explore design spaces and leverage guidance during workflow development.

Abstract:
Vision Language Models (VLMs) demonstrate promising chart comprehension capabilities. Yet, prior explorations of their visualization literacy have been limited to assessing their response correctness and fail to explore their internal reasoning. To address this gap, we adapted attention-guided class activation maps (AG-CAM) for VLMs, to visualize the influence and importance of input features (image and text) on model responses. Using this approach, we conducted an examination of four open-source (ChartGemma, Janus 1B and 7B, and LLaVA) and two closed-source (GPT-4o, Gemini) models comparing their performance and, for the open-source models, their AG-CAM results. Overall, we found that ChartGemma, a 3B parameter VLM fine-tuned for chart question-answering (QA), outperformed other open-source models and exhibited performance on par with significantly larger closed-source VLMs. We also found that VLMs exhibit spatial reasoning by accurately localizing key chart features, and semantic reasoning by associating visual elements with corresponding data values and query tokens. Our approach is the first to demonstrate the use of AG-CAM on early fusion VLM architectures, which are widely used, and for chart QA. We also show preliminary evidence that these results can align with human reasoning. Our promising open-source VLMs results pave the way for transparent and reproducible research in AI visualization literacy. Code and Supplemental Materials: https://osf.io/fp3rg

Abstract:
Data-driven news articles combine narrative storytelling with data visualizations to inform and influence public opinion on pressing societal issues. These articles often employ persuasive strategies, which are rhetorical techniques in narrative framing, visual rhetoric, or data presentation, to influence audience interpretation and opinion formation regarding information communication. While previous research has examined whether and when data visualizations persuade, the strategic choices made by persuaders remain largely unexplored. Addressing this gap, our work presents a taxonomy of persuasive strategies grounded in psychological theories and expert insights, categorizing 15 strategies across five dimensions: Credibility, Guided Interpretation, Reference-based Framing, Emotional Appeal, and Participation Invitation. To facilitate large-scale analysis, we curated a dataset of 936 data-driven news articles annotated with both persuasive strategies and their perceived effects. Leveraging this corpus, we developed a multimodal, multi-task learning model that jointly predicts the presence of persuasive strategies and their persuasive effects by incorporating both embedded (text and visualization) and explicit (visual narrative and psycholinguistic) features. Our evaluation demonstrates that our model outperforms state-of-the-art baselines in identifying persuasive strategies and measuring their effects.

Abstract:
This paper presents a novel approach for generating high-quality, cross-category 3D models from free-hand sketches with limited training data. We propose the first semi-supervised learning method to our knowledge for sketch-to-3D model conversion. Innovatively, we design a coarse-to-fine pipeline to perform the semi-supervised learning in the coarse stage and train a diffusion-based refiner to get a high-resolution 3D model. We designed a sketch-augmentation method for semi-supervised learning and integrated priors such as CLIP loss, shape prototypes, and adversarial loss to help generate high-quality results even with abstract and imprecise sketches. We also introduce an innovative procedural 3D generation method based on CAD code, which helps pre-train part of the network before fine-tuning with limited real data. Our approach, coupled with a specifically designed curriculum learning, allows us to generate high-quality 3D models across multiple categories with as few as 300 sketch-3D model pairs, marking a significant advancement over previous single-category approaches. In addition, we introduce the KO2D dataset, the largest collection of hand-drawn sketch-3D pairs to support further research in this area. As sketches are a far more intuitive and detailed way for users to express their unique ideas, we believe that this paper can move us closer to democratizing 3D content creation, enabling anyone to transform their ideas into 3D models effortlessly.

Abstract:
Cross-platform virtual environments allow users to experience interactive virtual places from a variety of peripherals, such as mobiles, computers or virtual reality headsets. These devices offer different interactivity and immersion levels, which shape the subjective experience of presence in the virtual space. Moreover, users may be provided with virtual representations, or avatars, to evolve in the environment. Controlling such avatars will likely induce a sense of embodiment and an identification toward these representations. When using cross-platform applications, users may have to transpose their experience across different devices, likely transporting their avatar and prior experience with them. In this context, understanding the impact of device transitions and prior experience may have over user experience constructs appears critical. We designed an experiment to explore their effects, which primarily unveiled that transitioning from high-immersion devices to low-immersion ones causes a significant drop in users’ sense of presence, embodiment and identification. These results are supposedly due to a comparison effect between devices emphasized by the strong expectations created by the first exposure in VR. Our study further unveiled a strong connection between the sense of embodiment and character identification demonstrated by statistical correlations. It finally confirmed that identification may persist across devices and exposures, shedding light on this avatar-related construct for research on cross-device realities.

Abstract:
Spatialized audio cues in Augmented Reality (AR) provide an alternative to visually dominated navigation systems, yet little is known about how device type and cue design jointly shape user experience and performance. This study presents a within-subject experiment comparing artificial versus nature sounds across two AR platforms: a smartphone and a mixed-reality headset. Thirty-four participants completed audio-only outdoor navigation tasks, with 27 included in the final analysis. Subjective measures (UEQ, NASA-TLX, custom items) and objective metrics (completion time, path efficiency) were collected. Results showed that nature sounds were rated significantly higher in efficiency, pleasantness, and overall navigation experience compared to artificial sounds, while device type had limited influence on subjective ratings. Participants with prior AR audio navigation experience completed tasks faster, suggesting training benefits. These findings highlight the importance of cue design and user adaptation for future AR navigation systems.

Abstract:
Emergency evacuation in VR must balance realism with clear guidance. However, most prior studies strengthen either sensory or social factors in isolation, leaving equal-geometry causal estimates of load versus crowd still lacking. We present SAFE-VR, a controlled testbed that orthogonally varies Environmental Load (low vs. high) and Crowd Dynamics (orderly vs. chaotic) while keeping layout, signage, and spawn constant. In a preregistered 2 x 2 between-subjects experiment (N=80), we analyzed time-to-exit, frame-coded behavior, presence, and workload to distangle sensory from social effects. Both factors impaired egress, with the High×Chaotic condition performing worst overall. For time-to-exit, effects were additive (no reliable Load×Crowd interaction); in contrast, Temporal demand showed a crossed interaction. High load increased effort, frustration, and object contacts; while chaotic flow increased route deviations, human contacts, and slowed exits. These patterns align with reliability-weighted cueing: as guidance becomes harder to perceive, participants may shift toward crowd-following, especially when flow is unstable. SAFE-VR thus delineates how load and crowd structure jointly shape route fidelity, collisions, and evacuation time, and highlights conditions where subjective time pressure diverges from objective delay.

Abstract:
The use of beyond-real interactions (BRI) expands the possibilities of virtual reality (VR) experiences by enabling novel capabilities, such as extending the range of manipulation or employing unusual body configurations. However, such interactions require users to adapt to novel sensorimotor mappings, which can lead to decreased performance and a poorer quality of user experience. Recently, proprioceptive suppression with noisy tendon electrical stimulation (n-TES) has been proposed to promote adaptation to BRI, though its effects have been reported to be limited. This study investigated the effects of proprioceptive suppression via n-TES on the adaptation process to BRI and the consequent changes in body schema, which have remained unexplored in prior research. Using a between-subjects design with 24 participants, we examined the effects of n-TES on a reaching task with Go-Go interaction, which extends virtual hands to enable users to reach farther than physically possible. Body schema changes were measured using a pointing task without visual feedback of the hand after adaptation to the Go-Go interaction. The results revealed that n-TES significantly altered the adaptation process by facilitating early learning and also enhanced body schema changes and aftereffects of adaptation under certain conditions. However, parameter fitting of the learning curves also suggested that n-TES might lead to a deterioration in the final level of performance achieved through adaptation. Our findings provide initial evidence that proprioceptive suppression by n-TES can contribute to facilitating sensorimotor adaptation within VR applications, while its practical utility may require adjusting stimulation to prevent potential side effects in the later stages of adaptation, specifically in trajectory smoothness.

Abstract:
Fear plays a crucial role in human behavior and psychological responses. However, accurately quantifying its intensity remains challenging. To address this, we proposed a labeling paradigm in virtual environments: while watching immersive VR videos, users continuously pressed the VR controller trigger to map their subjective fear intensity to a real-time continuous label. Immersive VR videos were utilized as the experimental scenes because they represent a controllable and easily standardized intervention in exposure therapy. To enhance labeling accuracy, haptic vibration cues were provided at preset trigger depth thresholds, serving as physical anchors that allowed subjects to distinguish between fear levels without visual confirmation. Building on this, we designed CTFear, which combined parallel convolutional neural networks with spatial and temporal Transformers and introduced a topology-aware spatial positional encoding to integrate cross-electrode information. Experimental results showed that CTFear achieved average F1 scores of 0.86,0.76, and 0.67 for the two-class, three-class, and four-class classification of fear intensity in cross-trial validation, respectively, and 0.80, 0.64, and 0.55 in cross-subject validation, outperforming various existing methods in multiple classification of fear intensity tasks. This indicated our method can effectively drive EEG decoding of fear intensity and offers a viable pathway for emotion monitoring and interaction in virtual environment.

Abstract:
This paper presents the first systematic investigation of how congruence in visual self-representation influences self-perception and behavior. We span a continuum from the physical self through avatars with graded self-similarity to clearly dissimilar avatars in virtual reality (VR). In a 1 ×4 within-user study, participants completed movement and quiz tasks in either physical reality or a digital twin environment in VR, where they embodied one of three avatars: a photorealistic self-similar avatar, a dissimilar same-gender avatar, or a dissimilar opposite-gender avatar. Subjective measures included presence, sense of embodiment, self-identification, and perceived change, and were complemented by an objective movement metric of behavioral change. Compared to physical reality, VR, even with a self-similar avatar, produced lower presence, a weaker sense of embodiment, and reduced self-identification, revealing a persistent gap in visual congruence. Within VR, self-similar avatars enhanced body ownership, self-location, and self-identification relative to dissimilar avatars. Conversely, dissimilar avatars produced measurable behavioral changes compared with self-similar ones. Gender cues, however, had little impact in gender-neutral tasks. Overall, the findings show that photorealistic self-similar avatars reinforce embodiment and self-identification. However, VR still falls short of achieving congruence with physical reality, underscoring key challenges for avatar realism and ecological validity.

Abstract:
The global prevalence of mental health disorders has created a substantial treatment gap. To support clinicians and increase access to care, researchers in the field of Artificial Intelligence (AI) and Virtual Reality (VR) have investigated technology-mediated psychotherapy for years. However, research about stakeholders' concerns and their readiness to use AI in psychotherapy remains scarce. This study focuses on a user-centered approach to accommodate patients' concerns and, based on the results, implement measures to foster self-disclosure and trust towards an embodied AI therapist in VR. First, we conducted an online study with mental health patients (N=152), which identified data autonomy and transparency as their primary ethical concerns. In a subsequent in-person VR study (N=90) we compared effects of increased data autonomy and transparency on self-disclosure and trust towards an embodied AI therapist. Results indicated that higher data autonomy led to greater self-disclosure, while transparency had no significant effect. Manipulating data autonomy and transparency did not affect perceived trust, though exploratory calculations revealed that women reported significantly higher trust levels than men. These findings illuminate patients' priorities and provide implications for technical designs for AI-driven mental health care.

Abstract:
Virtual reality (VR) enables immersive, tightly controlled experiments but complicates subjective measurement: moving participants out of the virtual environment (VE) to complete questionnaires changes the measurement context and can increase recall bias. Questionnaires embedded in the VE (_IN \textVRQs) enable immediate, time-efficient self-reports and repeated sampling, yet adoption is limited by concerns that _IN \textVRQs might bias primary experiential measures and by the implementation burden of usable interfaces. In this within-subject study, 43 participants completed post-task questionnaires using three representative 2D _IN \textVRQ interface designs: a body-anchored watch (direct touch), a world-anchored station (handheld pointer), and a world-anchored board at 5 m (laser pointer). Usability and task load differed significantly across interfaces, revealing clear design trade-offs. Interview feedback highlighted corresponding ergonomic themes. In contrast, ratings of presence and flow related to the constant task and VE showed no practically relevant differences across interfaces, supported by equivalence tests. Findings suggest that once basic usability requirements are met, interface choice is unlikely to meaningfully bias experiential measures in similar setups. This study contributes methodological guidance for implementing or selecting _IN \textVRQ tools to improve VR study quality, comparability, and reproducibility. Supplemental materials are available at osf.io/528yk (CC BY 4.0).

Abstract:
Hand hygiene (HH) is essential for preventing healthcare-associated infections, yet conventional monitoring approaches primarily capture event occurrence and provide limited insight into procedural quality, timing, and individualized feedback. To address these limitations, we present a real-time HH training and assessment framework that combines deep-learning-based WHO step recognition with a protocol-aware decision-support engine deployed on both a Desktop LED display and a Mixed Reality (MR) headset. The system supports two complementary modes: Concurrent-Feedback Coaching (CFC), which provides real-time sequence guidance and corrective prompts, and Uncued Retention Assessment (URA), which evaluates unguided execution and summarizes detected steps and errors. To support real-time deployment, we retrained and evaluated YOLOv12+MV, TimeSformer, and TSM on four heterogeneous HH datasets (PSCUH, Jurmala, METC, and Kaggle). While several YOLOv12 variants achieved strong recognition performance, the compact YOLOv12-n+MV model provided the most favorable accuracy-efficiency trade-off for deployment, achieving F1-scores of 0.99, 0.87, 0.72, and 0.58 across Kaggle, Jurmala, METC, and PSCUH, respectively, with low computational cost. This lightweight recognizer was integrated with temporal majority voting and a protocol-aware controller to support stable closed-loop interaction on Desktop and HoloLens 2. We evaluated the framework in a controlled mixed-methods study with N=20N=20 participants using a 2× 22×2 design (Desktop vs. MR; CFC vs. URA). Desktop yielded significantly faster, more temporally stable, and less error-prone HH performance than MR, whereas CFC reduced total completion time and URA reduced weighted mistake scores, indicating a speed-accuracy trade-off. Subjective results showed higher perceived usability for Desktop than for MR and for CFC than for URA. NASA-TLX further showed a higher workload for MR than Desktop across five subscales under counterbalancing, while URA increased perceived effort relative to CFC. Overall, these findings suggest that Desktop is more suitable when efficiency, stability, and lower workload are priorities, whereas CFC and URA can be selectively used to emphasize guided acquisition or independent recall within Desktop and MR HH training workflows.

Abstract:
Recent advances in generative AI have accelerated the deployment of virtual celebrity agents for commercial endorsements. However, little is known about how their visual style and vocal type, especially when these characteristics are generated via advanced AI reconstruction or voice cloning techniques, affect user trust, perceived realism, familiarity, and social presence. We extracted a representative clip of Sheldon Cooper from The Big Bang Theory as a baseline and generated multiple Sheldon virtual agents that varied in visual style (hand-sculpted, AI-reconstructed, or non-stylized) and vocal type (cloned or synthesized). A 3 × 2 within-subjects experiment (N = 30) revealed that visual style significantly affected user trust, perceived realism, familiarity, and social presence, while vocal type affected only perceived realism and familiarity. Comparisons between the virtual agents and the video baseline confirmed that even cutting-edge 3D modeling still differs significantly from authentic video representations. Behavioral data indicate that the relationship between interpersonal-distance variations and trust levels is not a simple linear one. These findings provide actionable guidance for designers leveraging generative AI to create trustworthy virtual avatars or agents and delineate new research avenues for virtual celebrity agents.

Abstract:
Families make key decisions about whether and how Virtual Reality (VR) enters children's lives, but their priorities rarely shape the research agendas and accountability structures that guide what is studied and built. We examine children's and families' views on (1) which knowledge is most needed to support decisions about children's home VR use, (2) which advances, tools, and content should be prioritized, and (3) which actors they see as responsible for addressing these priorities. We first conducted a secondary analysis of prior interviews with children and guardians to distill a structured set of family-perceived alternatives for home VR use. We then engaged 46 child-involved groups (84 children, 40 guardians) to collaboratively prioritize the perceived importance of these alternatives and assign responsibility to different actors. We found that families prioritized knowledge about VR's effects on brain development and behavior as the most needed, viewing it as the underlying mechanisms with downstream consequences for other effects. They treated safety monitoring as a prerequisite that should not depend solely on parental control, and ranked safety above affordability, comfort, and visual realism. This work contributes to the VR community by offering a children- and family-informed VR research agenda, along with insights into who families believe should be responsible for addressing these priorities.

Abstract:
Adaptive virtual reality (VR) applications are used in training and rehabilitation to provide personalized experience, through the adaptive logic that adjust the virtual environment based on the user's behavior. While users' behavior is influenced by affective and cognitive states (ACS), adaptive logic typically relies only on users' performance. First, this paper introduces a general method for the adaptation of VR applications that integrates ACS. We then provide an implementation of this model adapting difficulty with regards to both performance and mental workload. Finally, we present a user study (N=30) to compare our mental workload-based method (experimental) to a state-of-the-art adaptation relying on performance only (control). Results show that our adaptation method led to a decrease of 10.7% in mental workload, an increase of 22.8% in performance, and an overall better experience for most participants. These results were achieved without the participants' awareness of the adaptive logic of each condition. Taken together, our results promote the integration of ACS in adaptive VR to enhance users' experience and efficiency, and better fit the function of VR training applications.

Abstract:
In immersive VR applications, avatars shape users' identities and social presence. Personalization requires generating realistic outfits for avatars, often specified through a single reference image, while supporting seamless editing, adaptation to diverse avatars, and efficient rendering. Achieving these goals is challenging because VR avatars, even humanoid ones, exhibit substantial variations in body shapes and topologies. This inherent diversity makes it difficult to collect sufficient paired data, impeding the evolution of generalizable end-to-end image-to-garment models. To this end, we propose Tailor, a two-stage framework that dresses 3D humanoid avatars from a single reference image while preserving the wearing effects observed in the image. In the first stage, Tailor leverages a structured garment representation based on sewing patterns, enabling the network to predict garments in a low-dimensional, interpretable, and topology-independent space. In the second stage, Tailor performs instance-specific optimization to adapt the predicted sewing pattern to the avatar, ensuring consistent wearing effects across varying avatars. Furthermore, this framework also enables seamless garment editing, on-the-fly adaptation, and real-time rendering, making it particularly suitable for large-scale VR environments. Extensive experiments demonstrate that Tailor achieves results comparable to professional manual designs and produces garments that are both visually appealing and better aligned with reference styles than those generated by naive pattern-scaling baselines, as validated through human perceptual studies.

Abstract:
Augmented reality has shown promise as an assistive tool for mass casualty incident (MCI) triage to facilitate cognitive support during live response and to furnish immersive virtual MCI training. It has been shown that women and people of color experience less effective triaging in emergency room settings, yielding diminished healthcare outcomes. In the more chaotic and austere MCI event, triage efforts further determine which patients are left to expire and which get life-saving interventions before they arrive at the hospital. This work offers an exploratory within subjects investigation with actual first responders (N = 13) to examine preliminary patterns about whether diminished triage efficacy based on patient demographics can be captured during an augmented-reality virtual training simulation (1V) and/or within a ‘real’ patient training simulation (supported with AR triage interfaces) with patient actors (2R). While both conditions used a Hololens2, the 1V condition displayed AR virtual patients, whereas the ‘patient actor’ condition had real people scattered in a physical space needing triaged. We conduct ANOVAS and construct linear models to evaluate our data. Our exploratory analysis finds slight potential impacts of patient demographics in triage efficacy within the ‘real’ simulation, and more severe impacts in the virtual simulation. We further discuss implications of this work and explore underlying cognitive factors.

Abstract:
The Technology Acceptance Model (TAM) has been used extensively to understand technology adoption in the context of virtual reality (VR). The model includes external variables that are important drivers of attitudes towards adopting technology. In this mixed-methods study, we assessed the effects of cognitive engagement and individual dimensions of cognitive load (CL) on the attitudes driving the intention to use an AI-supported VR system for nursing students' patient management training: perceived usefulness (PU) and perceived ease of use (PEOU). The participants were a group of nursing students from a university in the Midwestern United States. We also explored interview data to understand the participants' perceptions about the resulting factors. The quantitative results showed that engagement and PEOU are significant predictors of PU, and so are frustration (one of the dimensions of CL) and engagement with PEOU. Interview data revealed that participants' frustrations did not always have a negative effect. They generally enhanced their engagement by making the scenarios feel realistic and valuable for skill development.

Abstract:
Avatars can serve as virtual representations of a user’s body, replicating their movements in real time within an immersive environment. When these movements are closely synchronized with those of the user, a sense of embodiment over the avatar is often experienced. Recent research shows that introducing spatial offsets between a user’s real movements and those of their avatar can alter motor behavior, with users aligning with the distorted avatar. While previous studies have typically focused on upper-limb movements toward visual targets, it remains unclear how users respond to distortions without such targets, particularly during lower-limb tasks where postural demands may impose constraints. This study examined whether users adjust their movements to follow their distorted avatar during upper- and lower-limb reaching without visual targets. Twenty-four participants performed reaching tasks while their avatar’s movements were spatially offset. Results revealed that participants counteracted the avatar’s distortion by moving in the opposite direction in both limb tasks. These findings suggest that, in the absence of explicit visual targets, participants treat spatial distortions as execution errors and prioritize internal movement goals over minimizing sensory conflict. This suggests that adaptation depends not only on a strong embodiment but also on whether the task performance permits reducing the mismatch between the user’s and the avatar’s movements.

Abstract:
The rapid growth and availability of event sequence data across domains requires effective analysis and exploration methods to facilitate decision-making. Visual analytics combines computational techniques with interactive visualizations, enabling the identification of patterns, anomalies, and attribute interactions. However, existing approaches frequently overlook the interplay between temporal and multivariate attributes. We introduce EventBox, a novel data representation and visual encoding approach for analyzing groups of events and their multivariate attributes. We have integrated EventBox into Sequen-C, a visual analytics system for the analysis of event sequences. To enable the agile creation of EventBoxes in Sequen-C, we have added user-driven transformations, including alignment, sorting, substitution and aggregation. To enhance analytical depth, we incorporate automatically generated statistical analyses, providing additional insight into the significance of attribute interactions. We evaluated our approach involving 21 participants (3 domain experts, 18 novice data analysts). We used the ICE-T framework to assess visualization value, user performance metrics completing a series of tasks, and interactive sessions with domain experts. We also present three case studies with real-world healthcare data demonstrating how EventBox and its integration into Sequen-C reveal meaningful patterns, anomalies, and insights. These results demonstrate that our work advances visual analytics by providing a flexible solution for exploring temporal and multivariate attributes in event sequences.

Abstract:
When designed deliberately, data visualizations can become powerful persuasive tools, influencing viewers' opinions, values, and actions. While researchers have begun studying this issue (e.g., to evaluate the effects of persuasive visualization), we argue that a fundamental mechanism of persuasion resides in rhetorical construction, a perspective inadequately addressed in current visualization research. To fill this gap, we present a focused analysis of octopus maps, a visual genre that has maintained persuasive power across centuries and achieved significant social impact. Employing rhetorical schema theory, we collected and analyzed 90 octopus maps spanning from the 19th century to contemporary times. We closely examined how octopus maps implement their persuasive intents and constructed a design space that reveals how visual metaphors are strategically constructed and what common rhetorical strategies are applied to components such as maps, octopus imagery, and text. Through the above analysis, we also uncover a set of interesting findings. For instance, contrary to the common perception that octopus maps are primarily a historical phenomenon, our research shows that they remain a lively design convention in today's digital age. Additionally, while most octopus maps stem from Western discourse that views the octopus as an evil symbol, some designs offer alternative interpretations, highlighting the dynamic nature of rhetoric across different sociocultural settings. Lastly, drawing from the lessons provided by octopus maps, we discuss the associated ethical concerns of persuasive visualization.

Abstract:
We investigate whether tactile charts support comprehension and learning of complex visualizations for blind and low-vision (BLV) individuals and contribute four tactile chart designs and an interview study. Visualizations are powerful tools for conveying data, yet BLV individuals typically can rely only on assistive technologies—primarily alternative texts—to access this information. Prior research shows the importance of mental models of chart types for interpreting these descriptions, yet BLV individuals have no means to build such a mental model based on images of visualizations. Tactile charts show promise to fill this gap in supporting the process of building mental models. Yet studies on tactile data representations mostly focus on simple chart types, and it is unclear whether they are also appropriate for more complex charts as would be found in scientific publications. Working with two BLV researchers, we designed 3D-printed tactile template charts with exploration instructions for four advanced chart types: UpSet plots, violin plots, clustered heatmaps, and faceted line charts. We then conducted an interview study with 12 BLV participants comparing whether using our tactile templates improves mental models and understanding of charts and whether this understanding translates to novel datasets experienced through alt texts. Thematic analysis shows that tactile models support chart type understanding and are the preferred learning method by BLV individuals. We also report participants' opinions on tactile chart design and their role in BLV education.

Abstract:
We contribute an in-depth analysis of the workflows and tensions arising from generative AI (genAI) use in biomedical visualization (BioMedVis). Although genAI affords facile production of aesthetic visuals for biological and medical content, the architecture of these tools fundamentally limits the accuracy and trustworthiness of the depicted information, from imaginary (or fanciful) molecules to alien anatomy. Through 17 interviews with a diverse group of practitioners and researchers, we qualitatively analyze the concerns and values driving genAI (dis)use for the visual representation of spatially-oriented biomedical data. We find that BioMedVis experts, both in roles as developers and designers, use genAI tools at different stages of their daily workflows and hold attitudes ranging from enthusiastic adopters to skeptical avoiders of genAI. In contrasting the current use and perspectives on genAI observed in our study with predictions towards genAI in the visualization pipeline from prior work, we refocus the discussion of genAI's effects on projects in visualization in the here and now with its respective opportunities and pitfalls for future visualization research. At a time when public trust in science is in jeopardy, we are reminded to first do no harm, not just in biomedical visualization but in science communication more broadly. Our observations reaffirm the necessity of human intervention for empathetic design and assessment of accurate scientific visuals. Supplemental study materials are available at https://osf.io/genaixbiomedvis/.

Abstract:
While many visualizations are built for domain users (e.g., biologists, machine learning developers), understanding how visualizations are used in the domain has long been a challenging task. Previous research has relied on either interviewing a limited number of domain users or reviewing relevant application papers in the visualization community, neither of which provides comprehensive insight into visualizations “in the wild” of a specific domain. This paper aims to fill this gap by examining the potential of using Large Language Models (LLM) to analyze visualization usage in domain literature. We use high-dimension (HD) data visualization in sing-cell transcriptomics as a test case, analyzing 1,203 papers that describe 2,056 HD visualizations with highly specialized domain terminologies (e.g., biomarkers, cell lineage). To facilitate this analysis, we introduce a human-in-the-loop LLM workflow that can effectively analyze a large collection of papers and translate domain-specific terminology into standardized data and task abstractions. Instead of relying solely on LLMs for end-to-end analysis, our workflow enhances analytical quality through 1) integrating image processing and traditional NLP methods to prepare well-structured inputs for three targeted LLM subtasks (i.e., translating domain terminology, summarizing analysis tasks, and performing categorization), and 2) establishing checkpoints for human involvement and validation throughout the process. The analysis results, validated with expert interviews and a test set, revealed three often overlooked aspects in HD visualization: trajectories in HD spaces, inter-cluster relationships, and dimension clustering. This research provides a stepping stone for future studies seeking to use LLMs to bridge the gap between visualization design and domain-specific usage.

Abstract:
Data-driven decision making has become a popular practice in science, industry, and public policy. Yet data alone, as an imperfect and partial representation of reality, is often insufficient to make good analysis decisions. Knowledge about the context of a dataset, its strengths and weaknesses, and its applicability for certain tasks is essential. Analysts are often not only familiar with the data itself, but also have data hunches about their analysis subject. In this work, we present an interview study with analysts from a wide range of domains and with varied expertise and experience, inquiring about the role of contextual knowledge. We provide insights into how data is insufficient in analysts' workflows and how they incorporate other sources of knowledge into their analysis. We analyzed how knowledge of data shaped their analysis outcome. Based on the results, we suggest design opportunities to better and more robustly consider both knowledge and data in analysis processes.

Abstract:
Parametric functions are an extremely efficient representation for 3D geometry, capable of compactly modelling highly complex objects. Once specified, parametric 3D objects allow for visualization at arbitrary levels of detail (LOD), at no additional memory cost, limited only by the amount of evaluated samples. However, mapping the sample evaluation to the hardware rendering pipelines of modern graphics processing units (GPUs) is not trivial. In this article, we propose a general method for efficient rendering of parametrically-defined 3D objects on modern hardware architectures. Our method adaptively analyzes, allocates and evaluates parametric function samples to produce high-quality renderings. Geometric precision can be modulated from few pixels down to sub-pixel level, enabling real-time frame rates of several 100 frames per second (FPS) for various parametric functions. We propose a dedicated LOD stage, which outputs patches of similar geometric detail to a subsequent rendering stage that uses either a hardware tessellation-based approach or performs point-based software rasterization. Our method requires neither preprocessing nor caching, and the proposed LOD mechanism is fast enough to run each frame. Hence, our approach also lends itself to animated parametric objects. We demonstrate the benefits of our method over a state-of-the-art spherical harmonics (SH) glyph rendering method and over classical LOD approaches, while showing its flexibility on a range of other demanding shapes.

Abstract:
We develop a probabilistic approach to understanding user preferences for adaptive placement of augmented reality (AR) interfaces in the physical environment through a series of user studies conducted using simulated desktop and virtual reality (VR) environments. From the first online crowdsourcing study and its validation in VR, we derived a set of potential factors behind user preferences for AR interface adaptation by assessing user-created layouts and analysing subjective user feedback. Building on this prior knowledge, we implemented a probabilistic optimisation system to generate adapted AR interfaces. Using generated layout pairs that prioritise different factors, we conducted a second online crowdsourcing study (N = 250) to elicit user preference rating data to quantify posterior probabilities for the weighting coefficients of the factors in the optimisation utility function. Overall, we found that the overall structures of layouts, such as shape and distribution, are more important to users than adapting to specific features of the environment, such as semantic associations between AR widgets and objects in the physical environments. We contribute a statistical model containing probabilistic distributions of different factors as a universal prior model that represents user preferences for AR interface placement that adapts to changing physical environments. Based on the results, we distil concrete guidelines for future adaptive AR interface systems regarding layout consistency, structure, and relationships between virtual widgets and physical objects.

Abstract:
Target selection is a fundamental interaction in virtual reality (VR). But the act of confirming a selection, such as a button press or pinch, can disturb the tracked pose and shift the intended target, which is referred to as the Heisenberg Effect. Prior research has mainly investigated controller input. However, it remains unclear how the effect manifests in the bare-hand input and how score-based techniques may mitigate the effect in different spatial variations. To fill the gap, we conduct a within-subject study to examine the Heisenberg Effect across two input modalities (i.e., controller and hand) and two selection mechanisms (i.e., direct and score-based). Our results show that hand input is more susceptible to the Heisenberg Effect, with direct selection more influenced by target width and score-based selection more sensitive to target density. Based on previous vote-oriented technique and our temporal analysis, we introduce weighted VOTE, a history-based intention accuracy model for target voting, that reweights recent interaction intent to counteract input disturbances. Our evaluation shows the method improves selection accuracy compared to baseline techniques. Finally, we discuss future directions for adaptive selection methods.

Abstract:
Integrating augmented reality (AR) into industrial assembly processes has great potential, yet most studies use simplified lab tasks and rarely benchmark AR against commercial systems, limiting their industrial relevance. To address this gap, we present a comparative study of immersive AR and non-immersive touchscreen-based assembly support systems on a manufacturing shop floor, with 16 participants assembling two industry-grade assets of different complexity. The results reveal that while AR's immersive capabilities excelled in spatial guidance for complex tasks, the simplicity and reliability of the touchscreen interface proved more effective for simple assemblies. We derive three design implications for deployment practice.

Abstract:
Integrated Virtual Reality (IVR) systems are central to avatar-mediated use cases in Virtual Reality (VR), reconstructing users' movements on avatars. They differ primarily in their tracking architectures, which determine how completely and accurately users' movements are captured and reconstructed on avatars. Many current IVR systems reduce user-worn hardware, trading reconstruction accuracy against cost and setup complexity, yet their impact on user experience and task performance across use cases remains underexplored. We compared three reduced user-worn IVR systems. Each system has distinct technical approaches: (1) Captury (markerless outside-in optical tracking), (2) Meta Movement SDK (markerless inside-out optical tracking), and (3) Vive Trackers (marker-based outside-in optical tracking with IMUs). In a 3x5 mixed-design, participants performed five tasks, simulating different use cases, to probe distinct aspects of these systems. No system consistently outperformed the others. Meta excelled in hand-based, fast-paced interactions, while Captury and Vive performed better in lower-body tasks and during full-body pose observation. These findings underscore the need to evaluate reduced user-worn IVR systems within the specific use case. We offer practical guidance for system selection based on use-case demands and released our tasks as an open-source, extensible framework to support future evaluations for selecting IVR systems.

Abstract:
This study investigates how interaction methods and spatial distance affect user performance and experience in virtual reality (VR) tasks. We compared the Go-Go technique with ray-based Laser selection across eight target distances (0.67× to 4× arm length) in a memory task to identify distance ranges that minimize motor and cognitive demands and support working memory. Results with 41 participants revealed a trade-off: Go-Go enhanced embodiment, while Laser yielded higher efficiency and lower workload. This trade-off was distance-dependent: Go-Go performed best at 1.2×-1.4×, and Laser at 1.4×-2.2×. Our results offer practical design guidelines for adaptive, distance-aware VR systems that align performance and subjective experience.

Abstract:
Effective collaboration in Virtual Reality depends on users' ability to accurately perceive the affordances of others within a shared space. This paper presents the first empirical investigation of how people judge others' reaching capabilities in Immersive Virtual Environments (IVEs). We examined whether observers can reliably estimate the maximum horizontal reach distance of virtual actors, how different actor representations influence these perceptual judgments, and whether calibration feedback can improve estimation accuracy. Participants (N=35) judged the reachability of objects for virtual actors across three visual fidelity conditions: high-fidelity, low-fidelity, and end-effector-only representations. Results demonstrate that users can successfully estimate others' horizontal reachability in IVEs, though with overestimation. Calibration through action feedback significantly improved estimation accuracy, with overestimation decreasing across trials. Visual fidelity showed no main effect on initial reachability judgments but significantly affected postcalibration performance, with low-fidelity representations demonstrating significantly better calibration outcomes compared to high-fidelity avatars, while end-effector-only representations showed no significant difference from either condition. The findings contribute to the social perception literature in Virtual Reality (VR) and to provide design guidelines for avatar systems in collaborative virtual environments, where spatial coordination and shared action planning are critical.

Abstract:
Deep Neural Network (DNN) evaluation is a classic topic in machine learning. Unlike existing methods that assess a DNN’s general performance on pre-collected real-world data, we propose an approach to exploring a DNN’s local performance on generated data. The core idea is to train a generative model (GM) on real-world data and visualize the DNN’s performance over the GM’s latent space (GLS). The GLS contains all generated samples (GSs) of the GM, capturing combinations of features and their variations present in the GM’s training set, thereby enhancing the comprehensiveness of testing. Moreover, the plane organizes all GSs into a continuous space with regular variations in features, enabling users to conduct cross-scale exploration to understand the DNN’s behavior across different levels of feature granularity. Our research focuses on two challenges in achieving and applying the idea. First, the GLS is unvisualizable due to its high dimensionality and unboundedness. We thus project the GLS onto a plane and visualize the DNN’s performance on this plane. The plane serves as a surrogate for the GLS, providing rich visual cues for locating regions containing GSs involving performance-relevant patterns. Second, the plane alone is insufficient for performance exploration. To overcome this, we develop a system based on it. The system comprises two main views that integrate interactions, such as panning and zooming, to enable users to observe any plane region and assess the DNN’s performance on the corresponding GSs. We also develop three auxiliary views to facilitate pattern exploration and to diagnose factors that lead to specific performance patterns. Experimental results demonstrate the usability and effectiveness of our approach.

Abstract:
Humans inherently connect certain colors with particular concepts in semantically meaningful ways that facilitate visual communication. These colors are known as semantically resonant colors. For instance, we associate “sky” and “ocean” with shades of blue, and “cherry” with red. In this paper, we investigate how language models, including Word2Vec, RoBERTa, GPT-4o mini and the vision language model CLIP generate and represent nuanced semantically resonant colors for diverse concepts. To achieve this, we utilized a large dataset of color names and concepts, tailored models for the structure of each language model, and developed an interactive web interface, Concept2Color, as a use case. Additionally, we conducted experiments and a detailed analysis to assess the ability of these models to generate meaningful colors. Through these experiments, we examined how factors such as model design, training data and context affect the color output. Our findings reveal the capabilities and limitations of language models in processing and generating semantically resonant colors for concepts, thus contributing insights into how they depict semantic color-concept connections. These insights have implications for data visualization, design, and human-computer interaction, where leveraging effective semantic color generation can enhance communication and user experience.

Abstract:
This work investigates the current research on in-situ visualisations for running: visualisations about data that are referred to during the running activity. We analyse 47 papers from 33 Human-Computer Interaction and Visualisation venues and identify six dimensions of a design space of in-situ running visualisations. Our analysis of this design space highlights an emerging trend: a shift from on-body, peripersonal visualisations (i.e., in the space within direct reach, such as visualisations on a smartwatch or a mobile phone display) towards extrapersonal displays (i.e., in the space beyond immediate reach, such as visualisations in immersive augmented reality displays) that integrate data in the runner's surrounding environment. We explore this opportunity by conducting a series of workshops with 10 active runners in total, eliciting design concepts for running visualisations and interactions beyond conventional 2D displays. We find that runners show a strong interest for visualisation designs that favour more context-aware, interactive, and unobtrusive experiences that seamlessly integrate with their run. These findings inform a set of design considerations for future immersive running visualisations and highlight directions for further research.

Abstract:
We conduct a deconstructive reading of a qualitative interview study with 17 visual data journalists from newsrooms across the globe. We borrow a deconstruction approach from literary critique to explore the instability of meaning in language and reveal implicit beliefs in words and ideas. Through our analysis we surface two sets of opposing implicit beliefs in visual data journalism: objectivity/subjectivity and humanism/mechanism. We contextualize these beliefs through a genealogical analysis, which brings deconstruction theory into practice by providing a historic backdrop for these opposing perspectives. Our analysis shows that these beliefs held within visual data journalism are not self-enclosed but rather a product of external societal forces and paradigm shifts over time. Through this work, we demonstrate how thinking with critical theories such as deconstruction and genealogy can reframe “success” in visual data storytelling and diversify visualization research outcomes. These efforts push the ways in which we as researchers produce domain knowledge to examine the sociotechnical issues of today's values towards datafication and data visualization. All supplemental materials for this work are available at osf.io/5fr48.

Abstract:
Although immersive technologies hold enormous potential for enhancing sports training, existing systems encounter notable limitations. While fully immersive virtual reality (VR) systems can address constraints of physical training setups such as limited space or equipment, full VR environments often lack rich, realistic, and multisensory feedback often required for effective sports training. To address this gap, this work investigates skill-based sport training, particularly, golf putting, across the reality-virtuality continuum. We implemented a mixed reality (MR) putting simulator that combines real equipment (putter and golf ball) with a virtual training environment. Three training conditions were realized: (i) real-world putting, constrained by limited physical space; (ii) a VR putting, in which a real putter and ball are used but represented virtually within the virtual environment (VE); (iii) an augmented virtuality (AV) putting, which reveals the real putter, ball, and users' limbs through a localized video-passthrough window, aiming to preserve users' sense of embodiment and presence while enabling simulation of an unlimited putting area. We conducted a pilot within-subjects experiment with 36 participants and found that the AV condition improved immediate putting performance, reduced task load, reduced perceived motion sickness, and increased user preference compared to the VR condition, while user's sense of presence remained comparable to VR training. These findings indicate that AV offers a pragmatic “sweet spot” on the virtuality continuum, balancing the tradeoff between sensory fidelity and training flexibility, offering valuable design insights for immersive sports training systems.

Abstract:
Embodied representations are known to influence user behavior in virtual environments through stereotypes associated with their appearance, a phenomenon known as the Proteus effect, or through perceived affordances. The latter concept refers to the suggested ways of interacting with the environment as perceived by the capabilities of the user’s virtual representation. The inherent ability of virtual reality to support non-anthropomorphic avatars, enabling new interaction modalities, paves the way for a deeper understanding of affordances in virtual environments. We designed an experiment to assess the effect of embodying a chimeric hand (crab claw), compared to a regular hand, on participants’ sense of embodiment, presence, character plausibility, mental workload, and behavior. Following a within-subject design, participants took part in a gamified task consisting of finding a key in a wrecked pirate ship. As the key could be hidden in containers of different durability, their strategy (opening or destroying objects) was used as an objective measure. Results indicated a high perceived usability and a low mental workload with both virtual limbs. Participants also performed hitting actions more frequently when using the biomimetic limb than when using the humanoid hand representation. A higher perceived change in the body schema was reported when embodying the crab claw, but a higher sense of body ownership and agency were observed with the anthropomorphic hand. Results revealed no significant difference in terms of plausibility.

Abstract:
Immersive virtual reality learning environments (IVRLEs) are increasingly used in medical education, yet the role of environmental fidelity—particularly scene design—remains underexplored. This study examines how varying levels of fidelity and contextualization affect motivational and cognitive outcomes. Eighty-seven medical students were randomly assigned to one of three scene conditions: a minimalistic “Blank Scene,” a “Reconstructed Classroom Scene”, or an “Inside-Human Scene”. All students used a custom-developed application to learn about embryonic heart development. We measured virtual presence, intrinsic motivation, cognitive load, learning outcomes, and usability. Results showed that scene design influenced virtual presence, selected aspects of intrinsic motivation, cognitive load, and learning outcomes. The Inside-Human Scene elicited higher physical and self-presence as well as higher comprehension scores compared to the Reconstructed Classroom Scene. The Reconstructed Classroom Scene was associated with higher extraneous cognitive load. Intrinsic cognitive load was rated higher in the Inside-Human Scene, while germane cognitive load did not differ between conditions. No significant differences were found for task performance or factual recall. Overall, the findings indicate that scene design in IVRLEs affects how learners engage with complex content and may support deeper understanding when perceptual and contextual properties are coherent, while visually detailed environments may increase extraneous cognitive demands without improving learning.

Abstract:
As extended reality (XR) systems become increasingly immersive and sensor-rich, they enable the collection of behavioral signals such as eye and body telemetry. These signals support personalized and responsive experiences and may also contain unique patterns that can be linked back to individuals. However, privacy mechanisms that naively pair unimodal mechanisms (e.g., independently apply privacy mechanisms for eye and body privatization) are often ineffective at preventing re-identification in practice. In this work, we systematically evaluate real-time privacy mechanisms for XR, both individually and in pair, across eye and body modalities. We assess privacy through re-identification rates and evaluate utility using numerical performance thresholds derived from existing literature to ensure real-time interaction requirements are met. We evaluated four eye and ten body mechanisms across multiple datasets, comprising up to 407 participants. Our results show that when carefully paired, multimodal mechanisms reduce re-identification rate from 80.3% to 26.3% in casual XR applications (e.g., VRChat and Job Simulator) and from 84.8% to 26.1 % in competitive XR applications (e.g., Beat Saber and Synth Riders), all while maintaining acceptable performance based on established thresholds. To facilitate adoption, we additionally release XR Privacy SDK, an open-source toolkit enabling developers to integrate the privacy mechanisms into XR applications for real-time use. These findings underscore the potential of modality-specific and context-aware privacy strategies for protecting behavioral data in XR environments.

Abstract:
Spatial skills, in particular mental rotation, are increasingly linked to STEM achievements and to computational thinking (CT) abilities. Traditional 2D spatial assessments face validity and equity issues from visual ambiguity, occlusion, and missing depth cues, requiring cognitively demanding 2D-to-3D interpretation. These artifacts might contribute to gender disparities, giving males an advantage that may reflect test bias rather than true ability differences. Immersive Virtual Reality (VR) seems to mitigate these issues by providing stereoscopic depth, multi-perspective viewing, realistic lighting, and embodied interaction. We compared spatial performance in VR and traditional 2D assessments using the Virtual Reality Mental Rotation Assessment (VRMRA) in a within-subjects study of 48 adolescents (ages 12–16). VR improved performance, yielding higher accuracy in Mental Rotation Test (MRT)-style tasks (+1.3 items, p <. 001) and faster completion in Purdue Spatial Visualization Test: Rotations (PSVT:R)-style tasks (−100s, p < .001), while PSVT:R accuracy did not differ significantly between 2D and VR. We observed that gender effects are task-specific: VR improved the performance of females most in MRT-style tasks, reversing the 2D male advantage, whereas males gained more in PSVT:R-style tasks. Spatial scores correlated positively with CT across both 2D and VR assessments, indicating a link to computational problem-solving. These results suggest that the assessment of spatial skills in VR produces distinct performance patterns with respect to traditional 2D assessment, highlighting the importance of immersive VR for the development of improved tools and its potential implications for inclusive STEM education and talent identification.

Abstract:
Immersive Virtual Environments (IVEs) incorporating tangible graspable objects are becoming more accessible for training and other applications. Visual and proprioceptive information has been shown to be important for accurate distance and size perception. However, previous research has shown that users overestimate the size of tangible objects in IVEs. To address this issue, researchers have demonstrated the use of calibration to alter users' perceptions. In this study, we examined the carryover effects of calibration on the perceived size of graspable dials in IVEs. Participants wore head-mounted displays and experienced two calibration perturbations: Divergent Plus (physical dials 10% larger than virtual) and Divergent Minus (physical dials smaller than virtual). These were compared to the baseline Convergent condition (equal diameters). The results showed significant effects of the calibration on the accuracy of diameter estimates. Overall, a higher overestimation of dial sizes for the haptics-only condition was observed as compared to the vision-only condition. Additionally, participants' estimations worsened when haptic information was added to the vision during calibration but improved when vision information was added to the haptics-only condition during calibration. These findings highlight the importance of calibration in improving the accuracy of size perception in IVEs and contribute to our understanding of how tangible graspable objects are perceived in immersive virtual environments.

Abstract:
Walking in immersive virtual reality (VR) environments is often disrupted by gait (i.e., walking patterns) instability, a challenge that is further exacerbated for individuals with mobility impairments. This study examines the effectiveness of multimodal feedback by integrating auditory, vibrotactile, and visual cues in improving walking performance within VR. A total of 68 participants, equally divided between those with mobility impairments and those without, completed walking tasks under multiple feedback conditions. Walking velocity was the primary performance metric, supplemented by subjective assessments of mental workload, fatigue, presence, usability, and simulator sickness. Results revealed that multimodal feedback significantly enhanced walking velocity compared to unimodal and bimodal conditions, with statistical analysis confirming strong improvements (p < .001). Participants also reported a favorable user experience under multimodal conditions despite slightly increased cognitive demand. These findings highlight the potential of integrated sensory feedback to mitigate gait disturbances in VR, promoting safer and more accessible locomotion for users with and without mobility impairments.

Abstract:
The inability of cochlear implant (CI) users to accurately localize sound sources poses daily safety risks and impedes social communication. While future augmented reality (AR) glasses are an ideal platform for delivering real-time visual and tactile assistive cues, their potential is hampered by a critical knowledge gap: how to design effective, non-distracting interfaces for this specific user group. To address this, we introduce a novel methodological approach using a virtual reality (VR) platform with high-fidelity color passthrough. This platform functions as an ecologically valid yet fully controllable testbed that faithfully simulates a future AR experience. Within this environment, we conducted a within-subjects study with 16 CI users to systematically evaluate three non-auditory cueing strategies: 1) Visual-Only, using color-coded peripheral widgets; 2) Tactile-Only, via a custom 8-channel vibrotactile array; and 3) a combined visuo-tactile Multimodal approach. We rigorously measured localization accuracy (absolute angular error), reaction time, front-back confusion rates, and subjective cognitive load (NASA-TLX). Our results reveal a critical performance-load trade-off: the multimodal strategy yielded the highest accuracy, reducing the mean error from 85.3° to 8.1°, but also imposed the greatest cognitive load. Conversely, the tactile-only strategy offered a compelling balance, significantly improving performance while being rated as the least cognitively demanding. From this analysis, we derive foundational insights and a set of evidence-based design guidelines to inform the creation of more effective and usable auditory awareness systems on future AR platforms.

Abstract:
Cursors and how they are presented significantly influence user experience in both VR and non-VR environments by shaping how users interact with and perceive interfaces. In traditional interfaces, cursors serve as a fundamental component for translating human movement into digital interactions, enhancing interaction accuracy, efficiency, and experience. The design and visibility of cursors can affect users’ ability to locate interactive elements and understand system feedback. In VR, cursor manipulation is more complex than in non-VR environments, as it can be controlled through hand, head, and gaze movements. With the arrival of the Apple Vision Pro, the use of gaze-controlled non-visible cursors has gained some prominence. However, there has been limited exploration of the effect of this type of cursor. This work presents a comprehensive study of the effects of cursor visibility (visible versus invisible) in gaze-based interactions within VR environments. Through two user studies, we investigate how cursor visibility impacts user performance and experience across different confirmation mechanisms and tasks. The first study focuses on selection tasks, examining the influence of target width, movement amplitude, and three common confirmation methods (air tap, blinking, and dwell). The second study explores pursuit tasks, analyzing cursor effects under varying movement speeds. Our findings reveal that cursor visibility significantly affects both objective performance metrics and subjective user preferences, but these effects vary depending on the confirmation mechanism used and task type. We propose eight design implications based on our empirical results to guide the future development of gaze-based interfaces in VR. These insights highlight the importance of tailoring cursor metaphors to specific interaction tasks and provide practical guidance for researchers and developers in optimizing VR user interfaces.

Abstract:
Users are often seated in the real environment, while their virtual avatars either remain standing stationary or move in virtual reality (VR). This creates posture inconsistencies between the real and virtual embodiment representations. The relationship between posture consistency in locomotion techniques and sense of presence in VR is still unclear. This study investigates how visual and somatosensory integration affects the sense of standing (SoSt) and the sense of self-motion (SoSm) when the sitting posture is varied slightly, including highlighting the importance of sitting posture for locomotion design in VR. The degree and occurrence of SoSt and SoSm were assessed by subjective experiments, and it was found that higher sitting and lower sitting postures present higher SoSt and lower SoSm, respectively. Invocation of SoSt also influences postural perception. Perception of travel distance varied according to the posture condition when identical visual flow was presented. The findings suggest that visual and somatosensory integration related to posture enhances SoSt and SoSm, and a sitting posture with a higher seating position is recommended in seated VR locomotion design.

Abstract:
Target-selection-based teleportation is one of the most widely used and researched travel techniques in immersive virtual environments, requiring the user to specify a target location with a selection ray before being transported there. This work explores the influence of the maximum reach of the parabolic selection ray, modeled by different emission velocities of the projectile motion equation, and compares the resulting teleportation performance to a straight ray as the baseline. In a user study with 60 participants, we asked participants to teleport as far as possible while still remaining within accuracy constraints to understand how the theoretical implications of the projectile motion equation apply to a realistic VR use case. We found that a projectile emission velocity of 14 \fracms14ms (resulting in a maximal reach of \text21.52 m21.52m) offered the best trade-off between selection distance and accuracy, with an inferior performance of the straight ray. Our results demonstrate the necessity to carefully set and report the projectile emission velocity in future work, as it was shown to directly influence user-selected distance, selection errors, and controller height during selection.

Abstract:
Advancements in accessibility technologies such as low-cost swell form printers or refreshable tactile displays promise to allow blind or low-vision (BLV) people to analyze data by transforming visual representations directly to tactile representations. However, it is possible that design guidelines derived from experiments on the visual perception system may not be suited for the tactile perception system. We investigate the potential mismatch between familiar visual encodings and tactile perception in an exploratory study into the strategies employed by BLV people to measure common graphical primitives converted to tactile representations. First, we replicate the Cleveland and McGill study on graphical perception using swell form printing with eleven BLV subjects. Then, we present results from a group interview in which we describe the strategies used by our subjects to read four common chart types. While our results suggest that familiar encodings based on visual perception studies can be useful in tactile graphics, our subjects also expressed a desire to use encodings designed explicitly for BLV people. Based on this study, we identify gaps between the perceptual expectations of common charts and the perceptual tools available in tactile perception. Then, we present a set of guidelines for the design of tactile graphics that accounts for these gaps. Supplemental material is available at https://osf.io/3nsfp/?view_only=7b7b8dcbae1d4c9a8bb4325053d13d9f.

Abstract:
While embodied conversational agents driven by Large Language Models (LLMs) are emerging as valuable tools for instruction in Augmented Reality (AR), a key challenge lies in crafting their personalities to optimize both instructional efficacy and user engagement. To address this, we present findings from a within-subjects experiment that compared task performance and user experience with a LEGO assembly task. Participants received guidance from a real human instructor and three virtual counterparts, whose Big Five personality profiles were designed to be: (1) a direct replica of the real human, (2) an “ideal” profile ba sed on pedagogical research, or (3) customized by the participant. Our results reveal a critical trade-off: instruction from the real expert resulted in superior task efficiency and clarity; however, among the virtual conditions, instructors with idealized or user-customized personalities fostered significantly higher levels of user engagement and social presence compared to the virtual replica. Crucially, allowing users to customize their instructor's persona led to the strongest preference for future interaction. These findings underscore that personality is a fundamental component in the design of AI-driven instructors, providing empirical evidence for navigating the balance between task-oriented guidance and personalized, socially resonant user experiences.

Abstract:
With the advancement of virtual reality (VR), human-robot interaction within remote collaboration contexts has emerged as a pivotal research area. However, the nuanced interplay between the user's perspective and the robot's level of automation, and its subsequent impact on the operator's sense of agency (SoA), remains underexplored. This study investigates how perspective and collaboration mode—differentiated by the robot's level of automation—affect the SoA in VR-based human-robot collaboration. We implemented a 2 (perspective: first-person view[FPV] vs. third-person view[TPV]) × 3 (collaboration mode: Direct Remote Control, Supervisory Cooperation, Spectator Mode) experimental design. SoA was assessed via a multimodal approach, measuring both the explicit judgement of agency (JoA) and the implicit feeling of agency (FoA). The results revealed a dissociation between these two components: explicit JoA was dictated by the collaboration mode, with active modes yielding a higher explicit JoA than the passive Spectator Mode. In contrast, implicit FoA exhibited a significant interaction effect. Under FPV, the influence of collaboration mode was pronounced, where Direct Remote Control elicited the strongest FoA; however, under TPV, the differences among collaboration modes disappeared. The relationship between cybersickness and SoA was also explored. These findings suggest that SoA is not a unitary concept but a multifaceted experience, co-shaped by the user's perspective and their role as either a decision-maker or an executor. Furthermore, we investigate the relationship between cybersickness and SoA. Our study offers novel insights for the design of VR-based human-robot systems.

Abstract:
Systems relying on ML have become ubiquitous, but so has biased behavior within them. Research shows that bias significantly affects stakeholders' trust in systems and how they use them. Further, stakeholders of different backgrounds view and trust the same systems differently. Thus, how ML models' behavior is explained plays a key role in comprehension and trust. We survey explainability visualizations, creating a taxonomy of design characteristics. We conduct user studies to evaluate five state-of-the-art visualization tools (LIME, SHAP, CP, Anchors, and ELl5) for model explainability, measuring how taxonomy characteristics affect comprehension, bias perception, and trust for non-expert ML users. Surprisingly, we find an inverse relationship between comprehension and trust: the better users understand the models, the less they trust them. We investigate the cause and find that this relationship is strongly mediated by bias perception: more comprehensible visualizations increase people's perception of bias, and increased bias perception reduces trust. We confirm this relationship is causal: Manipulating explainability visualizations to control comprehension, bias perception, and trust, we show that visualization design can significantly (p < 0.001) increase comprehension, increase perceived bias, and reduce trust. Conversely, reducing perceived model bias, either by improving model fairness or by adjusting visualization design, significantly increases trust even when comprehension remains high. Our work advances understanding of how comprehension affects trust and systematically investigates visualization's role in facilitating responsible ML applications.

Abstract:
Social virtual reality (VR) applications enable users to experience highly immersive mediated interactions and social experiences. Especially embodied nonverbal communication, like mediated social touch, is mostly unique to social VR and potentially a major factor in enabling fulfilling emotional experiences at a distance. In our within-subjects study (N = 28), we compared feedback from wearable vibrotactile haptic devices with two established visual presentation forms of social touch: clipping and pseudo-haptics. Our findings suggest that the addition of haptic feedback can elicit significantly higher emotional arousal and result in a stronger illusion of touch than visual presentations alone. In contrast to existing literature, there was no evidence of a difference between the immersive and emotional experience of clipping and pseudo-haptics in our sample. In addition, we also investigated whether simply wearing haptic feedback devices during a social touch interaction, i.e., without active haptic feedback, affects the illusion of touch and user experience. However, there was no evidence of a significant effect in the sample studied. A free copy of this paper, including the appendix, is available at https://doi.org/10.17605/OSF.IO/4FYKQ.

Abstract:
Virtual reality (VR) technology allows users to feel a virtual body as if it was their own (i.e., body ownership illusion). Previous studies have explored how the visual realism of the user’s virtual body influences the sense of body ownership in VR, focusing on the aspect of anthropomorphism. However, the effect of photorealism, another element that characterizes the visual realism of a virtual human, has not been systematically examined in the context of body ownership illusion. Therefore, we investigate the effect of the rendering style of virtual hands on the sense of body ownership, hypothesizing that the effect is affected by the rendering style of the virtual environment. In addition, we examine the effect of photorealism on presence (i.e., the sense of being there), as existing studies have offered inconsistent evidence. To this aim, we conducted a 3 × 3 mixed-design remote VR experiment (N = 117) that factored in the rendering styles of both the virtual hand and the environment and analyzed the subjective data from the questionnaire. The results suggest that neither the rendering style of virtual hands nor the consistency of the rendering style between the virtual hand and the environment influenced the sense of body ownership. Nevertheless, the more photorealistically the virtual environment was rendered, the stronger the presence. Our results indicate that different dimensions of virtual body realism affect the sense of body ownership differently, thereby highlighting the importance of applying a suitable classification of realism to understand the body ownership illusion for virtual bodies more in depth.

Abstract:
Amid the global decline in news credibility, immersive journalism in virtual reality (VR) offers opportunities to enhance trust and engagement through presence and interactivity. As the communicators of news, virtual news anchors play a key role in shaping audience perceptions. However, the mechanisms through which virtual anchors in VR elicit audience trust remain underexplored. Guided by the Stimulus-Organism-Response framework, this study investigates how the visual style and voiceover of VR news avatars influence perceived realism, and how these perceptions subsequently shape trust and news credibility. In a 2 × 2 within-subjects experiment (N = 24), participants viewed a virtual news anchor featuring either (1) a realistic or (2) a cartoon-like visual style, paired with either (1) an anthropomorphic or (2) a neutral voiceover. Results indicate that visual style exerted a dominant influence on perceived realism and trust in virtual anchors. By contrast, while voiceover type did not affect perceived realism, perceived voice realism significantly predicted trust, highlighting the nuanced role of vocal cues in VR news contexts. Notably, the interaction between visual and vocal realism failed to yield an additive effect on trust, and avatar trust did not significantly predict news credibility. Our findings shed further light on avatar trust in immersive news environments and provide design insights for the development of credible and trustworthy virtual anchors.