IEEE transactions on visualization and computer graphics最新文献_第5页

Voxel Grid-Based Depth Recovery From Monocular Structured Light. 基于体素网格的单目结构光深度恢复。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-16 DOI: 10.1109/TVCG.2026.3734755

Zhuohang Yu, Kai Wang, Juyong Zhang

{"title":"Voxel Grid-Based Depth Recovery From Monocular Structured Light.","authors":"Zhuohang Yu, Kai Wang, Juyong Zhang","doi":"10.1109/TVCG.2026.3734755","DOIUrl":"10.1109/TVCG.2026.3734755","url":null,"abstract":"We introduce an approach for depth estimation from images captured by monocular structured light systems, building upon the matching-free differentiable rendering paradigm with a key architectural contribution: replacing the implicit signed distance function (SDF) geometry representation with an explicit density voxel grid. Unlike methods that rely on image matching, our technique leverages the known pattern projection and camera setup to guide geometry optimization through differentiable volume rendering, without any explicit correspondence search. We isolate and optimize the geometry field by treating projected patterns as a known color field during rendering, thereby enabling faster convergence and higher-quality results. We further introduce a set of complementary losses-a patch-based SSIM loss, a distortion loss for unimodal density, and a surface-aware color loss-that are specifically designed for the voxel grid representation in monocular structured light settings. Experiments show that our method outperforms matching-based techniques in few-shot scenarios, reducing average depth error by over 30% on both synthetic and real-world scenes. Compared to prior matching-free methods using SDFs, our voxel grid-based framework avoids smoothness-induced artifacts at sharp edges, captures fine geometric discontinuities more faithfully, and trains nearly seven times faster. We provide a systematic analysis of the trade-offs between explicit voxel grids and implicit SDFs, clarifying why the explicit representation is better suited to monocular structured light depth recovery.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042947","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Depth-Adaptive Injection for Stylized Text-to-Motion. 用于程式化文本到动作的深度自适应注入。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-16 DOI: 10.1109/TVCG.2026.3734489

Shuaiying Hou, Hongyu Tao, Junjie Gao, Junheng Fang, Rong Zhang, Yongcong Zhang, Shiqi Zhou, Weiwei Xu

{"title":"Depth-Adaptive Injection for Stylized Text-to-Motion.","authors":"Shuaiying Hou, Hongyu Tao, Junjie Gao, Junheng Fang, Rong Zhang, Yongcong Zhang, Shiqi Zhou, Weiwei Xu","doi":"10.1109/TVCG.2026.3734489","DOIUrl":"10.1109/TVCG.2026.3734489","url":null,"abstract":"Discrete text-to-motion generation models are powerful tools for synthesizing diverse human motions, yet endowing them with controllable, high-fidelity stylistic variations remains challenging. Direct retraining or fine-tuning these models often disrupts the fragile discrete token distributions learned by masked generative models, leading to optimization instability, style-content entanglement, and degraded motion quality or text alignment. To address these limitations, we propose the Modular Injection Framework (MIF), a framework for conditional adaptation of masked discrete text-to-motion models. For optimization stability, MIF employs a continuous residual injection mechanism to fuse conditioning signals within the hidden states of the pretrained backbone. Crucially, through systematic ablation, we uncover a depth-aware structural behavior: assigning injection to different transformer layers with depth-adaptive weights introduces a beneficial architectural bias. This design effectively aids to strike better trade-offs between style representation and motion quality, allowing the network to modulate high-frequency stylistic details while preserving low-frequency semantic content. Furthermore, to explore the generality of MIF, we adopt a distribution-aware adaptation strategy to extend MIF to novel motion adaptation and cross-modal music-conditioned dance synthesis. Based on target-domain reconstruction and codebook utilization statistics, the strategy determines which pretrained components remain fixed and which are trainable. Extensive experiments demonstrate our method's superior stylized motion quality and robust cross-domain adaptability. Notably, our method also supports temporal and cross-task composition.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148984406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The Neurophysiology of Cybersickness: A Systematic Review of Neuromarkers, Neurostimulation, and Reporting Practices in Virtual Reality. 晕屏病的神经生理学：对虚拟现实中神经标记物、神经刺激和报告实践的系统回顾。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-15 DOI: 10.1109/TVCG.2026.3734375

Gael Van der Lee, Francois Cabestaing, Hakim Si-Mohammed

{"title":"The Neurophysiology of Cybersickness: A Systematic Review of Neuromarkers, Neurostimulation, and Reporting Practices in Virtual Reality.","authors":"Gael Van der Lee, Francois Cabestaing, Hakim Si-Mohammed","doi":"10.1109/TVCG.2026.3734375","DOIUrl":"10.1109/TVCG.2026.3734375","url":null,"abstract":"Visually induced motion sickness (VIMS) is a prevalent adverse response to virtual reality (VR) exposure, and its assessment has relied primarily on retrospective subjective questionnaires that cannot support continuous monitoring or real-time detection. This PRISMA-guided systematic review synthesizes 92 studies on the neurophysiology of VIMS in VR, retrieved from PubMed, Web of Science, and Scopus through March 2025, spanning electroencephalography, functional near-infrared spectroscopy, and non-invasive neurostimulation. The EEG literature converges on increased low-frequency (delta and theta) power as the most consistent spectral correlate of VIMS. Alpha-band findings have been inconsistent across prior work; meta-analytic comparison of hardware platforms reported here attributes a substantial portion of that inconsistency to VR display type, with head-mounted-display and screen-based studies differing systematically in the direction of alpha and beta responses. Beyond spectral power, the corpus documents event-related potential evidence of attentional and inhibitory-control costs, fNIRS hemodynamic correlates, resting-state EEG predictors of individual susceptibility, and early closed-loop mitigation systems driven by neurophysiological detection. A systematic examination of the classification and prediction literature reveals recurrent reporting gaps (undisclosed decision thresholds, absent discriminative metrics, and validation schemes that place the same participant's data in both training and test sets) that limit the interpretability of headline performance figures. Together with hardware-related heterogeneity, baseline-selection variability, and terminological inconsistency, these gaps constrain cross-study synthesis. The causal status of observed neural associations, cross-subject generalization of detection models, and the neurodynamics of adaptation and recovery remain open empirical questions.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042981","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Beyond Arm's Reach: Revisiting 3D Interaction Techniques for Virtual Display Management in Immersive Workspaces. 超越手臂的范围：在沉浸式工作空间中重新审视虚拟显示管理的3D交互技术。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-15 DOI: 10.1109/TVCG.2026.3734382

Dongyun Han, Donghoon Kim, Garrett Woodhouse, Isaac Cho

{"title":"Beyond Arm's Reach: Revisiting 3D Interaction Techniques for Virtual Display Management in Immersive Workspaces.","authors":"Dongyun Han, Donghoon Kim, Garrett Woodhouse, Isaac Cho","doi":"10.1109/TVCG.2026.3734382","DOIUrl":"10.1109/TVCG.2026.3734382","url":null,"abstract":"Immersive technologies can transform 3D space into an extended information space. Such environments allow users to access and organize multiple information units simultaneously, thereby supporting sensemaking processes. Prior studies have observed that users in VR often arrange virtual displays, individual information units, within arm's reach, resulting in circular or spherical layouts around the body. Although immersive environments can theoretically provide unlimited information space, users' effective working areas often remain constrained by physical limitations. This work aims to explore how different 3D interaction techniques influence user performance and spatial usability during analytic tasks. We present a formal user study evaluating three interaction techniques-Simple Hand, Ray Casting, and Arm Extension-in clustering tasks. Our results show that Ray Casting and Arm Extension supported broader spatial use and reduced physical strain in virtual display management compared to Simple Hand; however, broader spatial use does not necessarily guarantee better task performance. Based on the results, we further discuss their implications for spatial management and interaction design in immersive analytics.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

RecPoint: Zero-Shot Text-to-3D Point Cloud Generation Via Structure-Aware Local Semantic Rectification. RecPoint：通过结构感知的局部语义校正生成零射击文本到3d点云。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-15 DOI: 10.1109/TVCG.2026.3734327

Yan Zhou, Jinping Zhang, Bangzhen Liu, Dewang Ye, Qinghua Lu, Xuemiao Xu, Cheng Xu, Yuexia Zhou

{"title":"RecPoint: Zero-Shot Text-to-3D Point Cloud Generation Via Structure-Aware Local Semantic Rectification.","authors":"Yan Zhou, Jinping Zhang, Bangzhen Liu, Dewang Ye, Qinghua Lu, Xuemiao Xu, Cheng Xu, Yuexia Zhou","doi":"10.1109/TVCG.2026.3734327","DOIUrl":"10.1109/TVCG.2026.3734327","url":null,"abstract":"Zero-shot text-to-3D generation has garnered growing interest through the integration of vision-language foundation models and 3D generative frameworks. However, existing methods often overlook the preservation of local geometric semantics critical for producing semantically aligned and structurally coherent 3D point clouds. Here, we introduce RecPoint, a structure-aware framework for text-guided 3D shape generation that rectifies semantic drift by explicitly modeling local geometric relationships. Central to our method is a geometry-aware graph transformer that encodes point-wise adjacency into graph signals, enabling fine-grained feature learning and robust shape reconstruction. Complementing this, we propose a test-time locality-aware semantic rectification mechanism that bridges modality gaps by grounding textual features in visually similar, structurally aligned image embeddings. Without relying on paired text-shape data, RecPoint achieves superior fidelity and semantic alignment across standard benchmarks, outperforming previous state-of-the-art approaches.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148979105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Impact of localized diminished illumination for needle targeting procedures in optical see-through head-mounted displays. 光学透明头戴式显示器中针头瞄准过程中局部照明减弱的影响。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-14 DOI: 10.1109/TVCG.2026.3733895

Daniel R Allen, Eva Deligiannis, Charles C X Yuan, Joeana Cambranis Romero, Elvis C S Chen, Terry M Peters

{"title":"Impact of localized diminished illumination for needle targeting procedures in optical see-through head-mounted displays.","authors":"Daniel R Allen, Eva Deligiannis, Charles C X Yuan, Joeana Cambranis Romero, Elvis C S Chen, Terry M Peters","doi":"10.1109/TVCG.2026.3733895","DOIUrl":"10.1109/TVCG.2026.3733895","url":null,"abstract":"The use of Optical see-through (OST) Head- Mounted Displays (HMDs) in Surgical Navigation Systems (SNSs) has seen a stark increase in recent years due to the release of consumer-grade HMDs such as the Microsoft HoloLens 2. Despite the promise of in situ visualization methods in enhancing surgical navigation for minimally-invasive procedures, failure to address the incorrect occlusion cue will lead to errors in depth perception, eliciting potential procedural mistakes from the physician. Rendering methods alone fall short in addressing this limitation, as lighting from the physical world, in contrast to Video see-through devices, cannot be digitally manipulated in OSTs. We propose the incorporation of a \"black hole\" projector into an OST-based Augmented Reality (AR)-SNS to mitigate these perceptual issues. In the context of surgical needle guidance, we integrated electromagnetic tracking into the HoloLens 2 environment through a co-calibration process to enable accurate needle tracking with no user-centric line-of-sight limitations. Employing a generic foam-based phantom with a skin-mimicking layer on top, we conduct a user study (n = 16) to evaluate the efficacy of our projector approach for needle insertion procedures against standard surgical lighting conditions. Quantitative results from the study show that, while the BlackHole projector had no significant impact in pure depth estimation capabilities, when employed to a needle targeting task it significantly improved the mean relative targeting error from 4.58mm with the standard lighting conditions to 2.69mm with the BlackHole approach (p = 0.047). Qualitative results showed the projector effectively reduced the mental strain associated with both depth estimation and needle targeting tasks.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The Impact of Text Anchoring Methods on Cognitive Load and User Experience in Virtual Reality. 虚拟现实中文本锚定方法对认知负荷和用户体验的影响。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-14 DOI: 10.1109/TVCG.2026.3733549

Junqing Guo, Yan Liu, Tianyu Hou, Zhi Qiao, Wenyu Hu, Fanghao Song

{"title":"The Impact of Text Anchoring Methods on Cognitive Load and User Experience in Virtual Reality.","authors":"Junqing Guo, Yan Liu, Tianyu Hou, Zhi Qiao, Wenyu Hu, Fanghao Song","doi":"10.1109/TVCG.2026.3733549","DOIUrl":"10.1109/TVCG.2026.3733549","url":null,"abstract":"Textual information in virtual reality (VR) museums is an important source for users to understand exhibit content. The anchoring method of textual information may affect information visibility, exhibit occlusion, and visual attention allocation, thereby influencing users' cognitive load and user experience. However, there is still a lack of systematic research on how different anchoring methods affect users' information acquisition, reading processing, and experiential perception. Therefore, we compared 5 anchoring methods- attached, view-facing, pop-up, handheld, and fusion-in a VR museum, and collected behavioral performance, eye-tracking (E-T), and questionnaire data for analysis. The results showed near-ceiling accuracy across the different anchoring methods, with no significant differences in reaction time. The E-T results indicated that average saccade rate and average fixation time could reflect differences in cognitive load across anchoring methods, whereas pupil diameter showed no significant change. Subjective workload and user experience exhibited a stratification. Overall, the view-facing and handheld methods resulted in lower workload and better user experience, the fusion method performed at an intermediate level, and the attached and pop-up methods yielded higher subjective workload. These findings provide empirical evidence and design implications for selecting text anchoring methods and optimizing VR museum interfaces.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149042984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Particle Transformer: A Physics-Informed Transformer for Real-Time Fluid-Structure Interaction Simulation. 粒子变压器：实时流固耦合仿真的物理信息变压器。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-11 DOI: 10.1109/TVCG.2026.3732692

Qitong Wu, Bo Li, Shiguang Liu

{"title":"Particle Transformer: A Physics-Informed Transformer for Real-Time Fluid-Structure Interaction Simulation.","authors":"Qitong Wu, Bo Li, Shiguang Liu","doi":"10.1109/TVCG.2026.3732692","DOIUrl":"10.1109/TVCG.2026.3732692","url":null,"abstract":"Recently, the use of data-driven approaches to accelerate physical simulation has emerged as a prominent research frontier. However, existing methodologies often grapple with limitations such as constrained generalization capabilities, a lack of physical consistency, and performance that falls significantly short of real-time requirements. In this paper, we propose the Particle Transformer, a Physics-Informed Neural Network (PINN) designed to substitute the core computational steps of the Material Point Method (MPM), aiming to enhance efficiency while maintaining physical fidelity. Our proposed simulator takes the complete physical state of the particle system as input to predict the subsequent state of the system. This predicted output is then directly utilized as the input for the following time step without incorporating any supplementary numerical solvers, thereby enabling solver-free, low-latency simulation. Our approach fully leverages the contextual understanding capabilities of Transformers to learn complex physical laws. Simultaneously, it employs a hierarchical, block-based network architecture, allowing it to handle intricate scenarios involving large-scale particle systems. Furthermore, by exploiting the intrinsic properties of the MPM framework, we perform a single-step physical iteration on the predicted particles. This allows for the seamless formulation of physical loss terms, thereby constructing an integrated MPM-Transformer PINN framework. This design ensures that our simulator maintains high physical accuracy while achieving real-time performance. Experimental results demonstrate that the proposed method is effective across various fluid-solid coupling scenarios, supports real-time simulation, and exhibits robust stability. Moreover, the architecture is highly scalable and well-suited for integration into larger or more complex physical systems. This work identifies a promising direction toward scalable, real-time neural simulation of MPM-style coupled materials.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148946353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Understanding 3D Mid-Air Spatial Manipulation in Immersive Environments: A Taxonomy and Visual Representation. 理解三维空中空间操纵沉浸式环境：分类和视觉表现。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-10 DOI: 10.1109/TVCG.2026.3732921

Francielly Rodrigues, Run Yu, Doug A Bowman

{"title":"Understanding 3D Mid-Air Spatial Manipulation in Immersive Environments: A Taxonomy and Visual Representation.","authors":"Francielly Rodrigues, Run Yu, Doug A Bowman","doi":"10.1109/TVCG.2026.3732921","DOIUrl":"10.1109/TVCG.2026.3732921","url":null,"abstract":"Mid-air spatial manipulation in 3D virtual environments poses unique challenges, mainly due to the absence of physical constraints. To address these challenges, several manipulation techniques emerged. However, there is a lack of systematic understanding of the design space for such techniques due to limitations of existing taxonomies. We propose a novel framework for characterizing mid-air manipulation techniques based on five orthogonal components: Input Style, Frame of Reference, Manipulation Directness, Degrees of Freedom and Transfer Function Mappings. This taxonomy enables a more fine-grained classification that captures distinctions often overlooked or conflated in previous models. In addition, we introduce a visual representation designed to support the description, comparison, and design of 3D spatial manipulation techniques. To demonstrate its utility, we apply the framework to a set of representative techniques from the literature, showing how the visual representation facilitates spotting distinctions between its different operating modes, even when such differences are subtle. Our aim is for this framework to serve not only for systematic description and analysis, but also as a practical resource for comparing, identifying design trade-offs, and informing the development of new manipulation techniques.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148932031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Collaborative Disassembly for Multiple Users in Virtual Reality. 虚拟现实中多用户协同拆卸。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2026-09-09 DOI: 10.1109/TVCG.2026.3732564

Ziteng Wang, Jian Wu, Sichun Huang, Peike Wang, Peng Zhang, Min Zhang, Lili Wang

{"title":"Collaborative Disassembly for Multiple Users in Virtual Reality.","authors":"Ziteng Wang, Jian Wu, Sichun Huang, Peike Wang, Peng Zhang, Min Zhang, Lili Wang","doi":"10.1109/TVCG.2026.3732564","DOIUrl":"10.1109/TVCG.2026.3732564","url":null,"abstract":"Disassembly Sequence Planning (DSP) is critical in product lifecycle management but computationally demanding, with traditional and machine learning approaches facing challenges in scalability and lacking the real-time responsiveness required for interactive collaboration in virtual reality (VR). We present Dynamic Parallel Disassembly and Tasking (DPDT), a novel framework tailored for immersive collaboration. DPDT integrates a spatial-temporal feature-based heuristic to accelerate search prioritization and employs a dynamic reactivation strategy to resolve inter-part dependencies during parallel execution. Furthermore, a proximity-aware task assignment algorithm optimizes the translation of these plans into efficient multi-user instructions. We conduct extensive evaluations on a large-scale dataset against multiple baselines, showing that DPDT significantly outperforms state-of-the-art baselines. Specifically, for complex assemblies, our method achieves a computational speedup of up to 4.7 times while improving planning reliability by approximately 9.3 percentage points. These algorithmic gains translate directly into operational efficiency: a 24-participant user study in VR confirms that DPDT significantly reduces task completion time, physical-effort proxies, and NASA-TLX workload, validating its potential for effective collaborative maintenance training.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2026-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148922381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0