IEEE transactions on visualization and computer graphics最新文献_第6页

A Handheld Stiffness Display with a Programmable Spring and Electrostatic Clutches for Haptic Interaction in Virtual Reality. 手持刚度显示与可编程弹簧和静电离合器的触觉交互在虚拟现实。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616795

Ke Shi, Quan Xiong, Maozeng Zhang, Aiguo Song, Lifeng Zhu

{"title":"A Handheld Stiffness Display with a Programmable Spring and Electrostatic Clutches for Haptic Interaction in Virtual Reality.","authors":"Ke Shi, Quan Xiong, Maozeng Zhang, Aiguo Song, Lifeng Zhu","doi":"10.1109/TVCG.2025.3616795","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616795","url":null,"abstract":"Handheld haptic devices often face challenges in delivering stiffness feedback with both high force output and good backdrivability, especially under practical constraints on power consumption, size, and weight. These difficulties stem from the inherent performance limitations of conventional actuation mechanisms. To address this issue, we propose a lightweight, low-power handheld device that provides wide-range stiffness feedback through a novel dual actuation design composed of two key components. A programmable spring (PS), implemented via an adjustable lever arm, enables tunable physical stiffness. Two electrostatic clutches (ECs) are integrated to compensate for the inherent limitations of PS-based interactions in stiffness display range, rendered object size, and free motion capability. The feedback force arises passively from the reaction of the PS and ECs to user input, effectively lowering both power consumption and actuator torque demands. A fully integrated prototype was developed, incorporating wireless communication, control, and power modules. The results of the evaluation experiments and user studies demonstrate that the device effectively renders stiffness across the full range, from free motion to full rigidity, and delivers more realistic elastic feedback compared to conventional electric motor-based systems.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214950","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

Seeing What Matters: Attentional (Mis-)Alignment Between Humans and AI in VR-Simulated Prediction of Driving Accidents. 看到什么是重要的：在vr模拟的驾驶事故预测中，人类和人工智能之间的注意力（错误）一致。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616811

Hoe Sung Ryu, Uijong Ju, Christian Wallraven

{"title":"Seeing What Matters: Attentional (Mis-)Alignment Between Humans and AI in VR-Simulated Prediction of Driving Accidents.","authors":"Hoe Sung Ryu, Uijong Ju, Christian Wallraven","doi":"10.1109/TVCG.2025.3616811","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616811","url":null,"abstract":"This study explores how human and AI visual attention differ in a short-term prediction task, particularly in the moments before an accident is about to happen. Since real-world studies of this kind would pose ethical and safety risks, we employed virtual reality (VR) to simulate an accident scenario. In the scenario, the driver approaches a fork in the road, knowing that one path would lead off a cliff crashing the car fatally-as the fork comes closer, the other, safe, path is suddenly blocked by trees, forcing the driver to make a split-second decision where to go. A total of $N = 71$ drivers completed the task, and we asked another $N = 30$ observers to watch short video clips leading up to the final event and to predict which way the driver would take. We then compared both prediction accuracy as well as attention patterns-how focus is distributed across objects-with AI systems, including vision language models (VLMs) and vision-only models. We found that overall, prediction performance increased as the accident time point approached; interestingly, humans fared better than AI systems overall except for the final time period just before the event. We also found that humans adapted their attention dynamically, shifting focus to important scene elements before an event, whereas AI attention remained static, overlooking key details of the scene. Importantly, as the accident time point approached, human-AI attentional alignment decreased, even though both types of models improved in prediction accuracy. Despite distinct temporal trajectories-vision-only models declining from an early advantage and VLMs peaking in the middle-both models achieved low to zero alignment with human attention. These findings highlight a critical dissociation: AI models make accurate predictions, but rely on visual strategies diverging from human processing, underscoring a gap between explainability and task performance.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214931","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

Extended Reality Check: Evaluating XR Prototyping for Human-Robot Interaction in Contact-Intensive Tasks. 扩展现实检验：在接触密集型任务中评估人机交互的XR原型。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616753

Tonia Mielke, Mareen Allgaier, Christian Hansen, Florian Heinrich

{"title":"Extended Reality Check: Evaluating XR Prototyping for Human-Robot Interaction in Contact-Intensive Tasks.","authors":"Tonia Mielke, Mareen Allgaier, Christian Hansen, Florian Heinrich","doi":"10.1109/TVCG.2025.3616753","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616753","url":null,"abstract":"Extended Reality (XR) has the potential to improve efficiency and safety in the user-centered development of human-robot interaction. However, the validity of using XR prototyping for user studies for contact-intensive robotic tasks remains underexplored. These in-contact tasks are particularly relevant due to challenges arising from indirect force perception in robot control. Therefore, in this work, we investigate a representative example of such a task: robotic ultrasound. A user study was conducted to assess the transferability of results from a simulated user study to real-world conditions, comparing two force-assistance approaches. The XR simulation replicates the physical study set-up employing a virtual robotic arm, its control interface, ultrasound imaging, and two force-assistance methods: automation and force visualization. Our results indicate that while differences in force deviation, perceived workload, and trust emerge between real and simulated setups, the overall findings remain consistent. Specifically, partial automation of robot control improves performance and trust while reducing workload, and visual feedback decreases force deviation in both real and simulated conditions. These findings highlight the potential of XR for comparative studies, even in complex robotic tasks.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214892","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

LeOp-GS: Learned Optimizer with Dynamic Gradient Update for Sparse-View 3DGS. LeOp-GS：学习优化器与动态梯度更新稀疏视图3DGS。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616156

Xinyu Lei, Xuan Wang, Longjun Liu, Haoteng Li, Haonan Zhang

{"title":"LeOp-GS: Learned Optimizer with Dynamic Gradient Update for Sparse-View 3DGS.","authors":"Xinyu Lei, Xuan Wang, Longjun Liu, Haoteng Li, Haonan Zhang","doi":"10.1109/TVCG.2025.3616156","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616156","url":null,"abstract":"3D Gaussian Splatting (3DGS) achieves remarkable speed and performance in novel view synthesis but suffers from overfitting and degraded reconstruction when handling sparse-view inputs. This paper innovatively addresses this challenge from a learning-to-optimize perspective by leveraging a learned optimizer (i.e., a multi-layer perceptron, MLP) to update the relevant parameters of 3DGS during the optimization process. Evidently, using a single MLP to handle all optimization variables, whose numbers may even vary during the optimization process, is impossible. Therefore, we present a point-wise position-aware optimizer that updates the parameters for each 3DGS point individually. Specifically, it takes the point coordinates and corresponding parameter values as input to predict the updates, thereby allowing efficient and adaptive optimization. In the case of sparse view modeling, the learned optimizer imposes position-aware constraints on the parameter updates during optimization. This effectively encourages the relevant parameters to converge stably to better solutions. To update the optimizer's parameters, we propose a dynamic gradient update strategy based on spatial perturbation and weighted fusion, enabling the optimizer to capture broader contextual information. Experiments demonstrate that our method effectively addresses the problem of modeling 3DGS from sparse training views, achieving state-of-the-art results across multiple datasets.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214895","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

Improving Mid-Air Sketching in Room-Scale Virtual Reality with Dynamic Color-to-Depth and Opacity Cues. 改进半空素描在房间规模的虚拟现实与动态颜色到深度和不透明度线索。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616867

Samantha Monty, Dennis Mevisen, Marc Erich Latoschik

{"title":"Improving Mid-Air Sketching in Room-Scale Virtual Reality with Dynamic Color-to-Depth and Opacity Cues.","authors":"Samantha Monty, Dennis Mevisen, Marc Erich Latoschik","doi":"10.1109/TVCG.2025.3616867","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616867","url":null,"abstract":"Immersive 3D mid-air sketching systems liberate users from the confines of traditional 2D sketching canvases. However, complications from perceptual challenges in Virtual Reality (VR), combined with the ergonomic and cognitive challenges of sketching in all three dimensions in mid-air lower the accuracy and aesthetic quality of 3D sketches. This paper explores how color-to-depth and opacity cues support users to create and perceive freehand, 3D strokes in room-scale sketching, unlocking a full 360°of freedom for creation. We implemented three graphic depth shader cues modifying the (1) alpha, (2) hue, and (3) value levels of a single color to dynamically adjust the color and transparency of meshes relative to their depth from the user. We investigated how these depth cues influence sketch efficiency, sketch quality, and total sketch experience with 24 participants in a comparative, counterbalanced, 4 × 1 within-subjects user study. First, with our graphic depth shader cues we could successfully transfer results of prior research in seated sketching tasks to room-scale scenarios. Our color-to-depth cues improved the similarity of sketches to target models. This highlights the usefulness of the color-to-depth approach even for the increased range of motion and depth in room-scale sketching. Second, our shaders assisted participants to complete tasks faster, spend a greater percentage of task time sketching, reduced the feeling of mental tiredness and improved the feeling of sketch efficiency in room-scale sketching. We discuss these findings and share our insights and conclusions to advance the research on improving spatial cognition in immersive sketching systems.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214913","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

Experiencing Immersive Virtual Nature for Well-Being, Restoration, Performance, and Nature Connectedness: a Scoping Review. 体验沉浸式虚拟自然的福祉，恢复，性能和自然连接：范围审查。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616762

Jeewoo Kim, Svara Patel, Hyeongil Nam, Janghee Cho, Kangsoo Kim

{"title":"Experiencing Immersive Virtual Nature for Well-Being, Restoration, Performance, and Nature Connectedness: a Scoping Review.","authors":"Jeewoo Kim, Svara Patel, Hyeongil Nam, Janghee Cho, Kangsoo Kim","doi":"10.1109/TVCG.2025.3616762","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616762","url":null,"abstract":"This paper presents a scoping review of immersive virtual nature experiences delivered via head-mounted displays (HMDs) and their role in promoting well-being, psychological restoration, cognitive performance, and nature connectedness. As access to natural environments becomes increasingly constrained by urbanization, technological lifestyles, and environmental change, immersive technologies offer a scalable and accessible alternative for engaging with nature. Guided by three core research questions, this review explores how HMD-mediated immersive technologies have been used to promote nature connectedness and well-being, what trends and outcomes have been observed across applications, and what methodological gaps or limitations exist in this growing body of work. Fifty-five peer-reviewed studies were analyzed and categorized into six key implication areas: emotional well-being, stress reduction, cognitive performance, attention recovery, restorative benefits, and nature connectedness. The review identifies immersive virtual nature as a promising application of extended reality (XR) technologies, with potential across healthcare, education, and daily life, while also emphasizing the need for more consistent methodologies and long-term research.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214884","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

Exploring the Effects of Augmented Reality Guidance Position within a Body-Fixed Coordinate System on Pedestrian Navigation. 在人体固定坐标系下增强现实导引位置对行人导航的影响研究。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616773

Shunbo Wang, Qing Xu, Klaus Schoeffmann

{"title":"Exploring the Effects of Augmented Reality Guidance Position within a Body-Fixed Coordinate System on Pedestrian Navigation.","authors":"Shunbo Wang, Qing Xu, Klaus Schoeffmann","doi":"10.1109/TVCG.2025.3616773","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616773","url":null,"abstract":"AR head-mounted displays (HMDs) facilitate pedestrian navigation by integrating AR guidance into users' field of view (FOV). Displaying AR guidance using a body-fixed coordinate system has the potential to further leverage this integration by enabling users to control when the guidance appears in their FOV. However, it remains unclear how to effectively position AR guidance within this coordinate system during pedestrian navigation. Therefore, we explored the effects of three AR guidance positions (top, middle, and bottom) within a body-fixed coordinate system on pedestrian navigation in a virtual environment. Our results showed that AR guidance position significantly influenced eye movements, walking behaviors, and subjective evaluations. The top position resulted in the shortest duration of fixations on the guidance compared to the middle and bottom positions, and lower mental demand than the bottom position. The middle position had the smallest rate of vertical eye movement during gaze shifts between the guidance and the environment, and the smallest relative difference in walking speed between fixations on the guidance and the environment compared to the top and bottom positions. The bottom position led to the shortest duration and smallest amplitude of gaze shifts between the guidance and the environment compared to the top and middle positions, and lower frustration than the top position. Based on these findings, we offer design implications for AR guidance positioning within a body-fixed coordinate system during pedestrian navigation.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214908","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

Application of Transitional Mixed Reality Interfaces: A Co-design Study with Flood-prone Communities. 过渡混合现实界面的应用：与洪水易发社区的协同设计研究。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616755

Zhiling Jie, Geert Lugtenberg, Renjie Zhang, Armin Teubert, Makoto Fujisawa, Hideaki Uchiyama, Kiyoshi Kiyokawa, Isidro Butaslac, Taishi Sawabe, Hirokazu Kato

{"title":"Application of Transitional Mixed Reality Interfaces: A Co-design Study with Flood-prone Communities.","authors":"Zhiling Jie, Geert Lugtenberg, Renjie Zhang, Armin Teubert, Makoto Fujisawa, Hideaki Uchiyama, Kiyoshi Kiyokawa, Isidro Butaslac, Taishi Sawabe, Hirokazu Kato","doi":"10.1109/TVCG.2025.3616755","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616755","url":null,"abstract":"Flood risk communication in disaster-prone communities often relies on traditional tools (e.g., paper and browser-based hazard/flood maps) that struggle to engage community stakeholders and reflect intuitive flood situations. In this paper, we applied the transitional mixed reality (MR) interface concept from pioneering work and extended it for flood risk communication scenarios through co-design with community stakeholders to help vulnerable residents understand flood risk and facilitate preparedness. Starting with an initial transitional MR prototype, we conducted three iterative workshops - each dedicated to device usability, visualization techniques, and interaction methods. We collaborated with diverse community stakeholders in flood-prone areas, collecting feedback to refine the system according to community needs. Our preliminary evaluation indicates that this co-designed system significantly improves user understanding and engagement compared to traditional tools, though some older residents faced usability challenges. We detailed this iterative co-design process, critical insights and design implications, offering our work as a practical case of mixed reality application in strengthening flood risk communication. We also discuss the system's potential to support community-driven collaboration in flood preparedness.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145214927","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

Source-Free Model Adaptation for Unsupervised 3D Object Retrieval. 无监督三维对象检索的无源模型自适应。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3617082

Dan Song, Yiyao Wu, Yuting Ling, Diqiong Jiang, Yao Jin, Ruofeng Tong

{"title":"Source-Free Model Adaptation for Unsupervised 3D Object Retrieval.","authors":"Dan Song, Yiyao Wu, Yuting Ling, Diqiong Jiang, Yao Jin, Ruofeng Tong","doi":"10.1109/TVCG.2025.3617082","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3617082","url":null,"abstract":"With the explosive growth of 3D objects yet expensive annotation costs, unsupervised 3D object retrieval has become a popular but challenging research area. Existing labeled resources have been utilized to aid this task via transfer learning, which aligns the distribution of unlabeled data with the source one. However, the labeled resource are not always accessible due to the privacy disputes, limited computational capacity and other thorny restrictions. Therefore, we propose source-free model adaptation task for unsupervised 3D object management, which utilizes a pre-trained model to boost the performance with no access to source data and labels. Specifically, we compute representative prototypes to assume the source feature distribution, and design a bidirectional cumulative confidence-based adaptation strategy to adaptively align unlabeled samples towards prototypes. Subsequently, a dual-model distillation mechanism is proposed to generate source hypothesis for remedying the absence of ground-truth labels. The experiments on a cross-domain retrieval benchmark NTU-PSB (PSB-NTU) and a cross-modality retrieval benchmark MI3DOR also demonstrate the superiority of the proposed method even without access to raw data. Code is available at: https://github.com/Wyyspace1203/MA.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145215006","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

SGSG: Stroke-Guided Scene Graph Generation. 描边引导场景图生成。

IF 6.5

IEEE transactions on visualization and computer graphics Pub Date : 2025-10-02 DOI: 10.1109/TVCG.2025.3616751

Qixiang Ma, Runze Fan, Lizhi Zhao, Jian Wu, Sio-Kei Im, Lili Wang

{"title":"SGSG: Stroke-Guided Scene Graph Generation.","authors":"Qixiang Ma, Runze Fan, Lizhi Zhao, Jian Wu, Sio-Kei Im, Lili Wang","doi":"10.1109/TVCG.2025.3616751","DOIUrl":"https://doi.org/10.1109/TVCG.2025.3616751","url":null,"abstract":"3D scene graph generation is essential for spatial computing in Extended Reality (XR), providing structured semantics for task planning and intelligent perception. However, unlike instance-segmentation-driven setups, generating semantic scene graphs still suffer from limited accuracy due to coarse and noisy point cloud data typically acquired in practice, and from the lack of interactive strategies to incorporate users, spatialized and intuitive guidance. We identify three key challenges: designing controllable interaction forms, involving guidance in inference, and generalizing from local corrections. To address these, we propose SGSG, a Stroke-Guided Scene Graph generation method that enables users to interactively refine 3D semantic relationships and improve predictions in real time. We propose three types of strokes and a lightweight SGstrokes dataset tailored for this modality. Our model integrates stroke guidance representation and injection for spatio-temporal feature learning and reasoning correction, along with intervention losses that combine consistency-repulsive and geometry-sensitive constraints to enhance accuracy and generalization. Experiments and the user study show that SGSG outperforms state-of-the-art methods 3DSSG and SGFN in overall accuracy and precision, surpasses JointSSG in predicate-level metrics, and reduces task load across all control conditions, establishing SGSG as a new benchmark for interactive 3D scene graph generation and semantic understanding in XR. Implementation resources are available at: https://github.com/Sycamore-Ma/SGSG-runtime.","PeriodicalId":94035,"journal":{"name":"IEEE transactions on visualization and computer graphics","volume":"PP ","pages":""},"PeriodicalIF":6.5,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145215038","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