Melania Prieto-Martín, Marc Comino-Trinidad, Dan Casas
{"title":"Detecting anomalies in dense 3D crowds","authors":"Melania Prieto-Martín, Marc Comino-Trinidad, Dan Casas","doi":"10.1016/j.cag.2025.104267","DOIUrl":"10.1016/j.cag.2025.104267","url":null,"abstract":"<div><div>Estimating the behavior of dense 3D crowds is crucial for applications in security, surveillance, and planning. Detecting events in such crowds from a single video, the most common scenario, is challenging due to ambiguities, occlusions, and complex human behavior. To address this, we propose a method that overlays pixel-based labels on video data to highlight anomalies in dense 3D crowds movement. Our key contribution is a data-driven, image-based model trained on features derived from 3D virtual crowd animations of articulated characters that mimic real crowds at a micro-level. By using training data based on captured dense crowd trajectories and realistic 3D motions, we can analyze and detect anomalies in complex real-world scenarios. Additionally, while acquiring ground-truth data from diverse viewpoints is difficult in real-world settings, our virtual simulator allows rendering scenes from multiple perspectives, enabling the training of models robust to viewpoint variations. We demonstrate qualitatively and quantitatively that our method can detect anomalies in much denser crowds than existing methods.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104267"},"PeriodicalIF":2.5,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144313425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"On-site single image SVBRDF reconstruction with active planar lighting","authors":"Lianghao Zhang, Ruya Sun, Li Wang, Fangzhou Gao, Zixuan Wang, Jiawan Zhang","doi":"10.1016/j.cag.2025.104268","DOIUrl":"10.1016/j.cag.2025.104268","url":null,"abstract":"<div><div>Recovering the spatially-varying bidirectional reflectance distribution function (SVBRDF) from a single image in uncontrolled environments is challenging while essential for various applications. In this paper, we address this highly ill-posed problem using a convenient capture setup and a carefully designed reconstruction framework. Our proposed setup, which incorporates an active extended light source and a mirror hemisphere, is easy to implement for even common users and requires no careful calibration. These devices can simultaneously capture uncontrolled lighting, real active lighting patterns, and material appearance in a single image. Based on all captured information, we solve the reconstruction problem by designing lighting clues that are semantically aligned with the input image to aid the network in understanding the captured lighting. We further embed lighting clue generation into the network’s forward pass by introducing real-time rendering. This allows the network to render accurate lighting clues based on predicted normal variations while jointly learning to reconstruct high-quality SVBRDF. Moreover, we also use captured lighting patterns to model noises of pattern display in real scenes, which significantly increases the robustness of our methods on real data. With these innovations, our method demonstrates clear improvements over previous approaches on both synthetic and real-world data.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104268"},"PeriodicalIF":2.5,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144329524","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"VectorMamba: Enhancing point cloud analysis through vector representations and state space modeling","authors":"Zhicheng Wen","doi":"10.1016/j.cag.2025.104255","DOIUrl":"10.1016/j.cag.2025.104255","url":null,"abstract":"<div><div>Point cloud data, despite its widespread adoption, poses significant challenges due to its sparsity and irregularity. Existing methods excel in capturing complex point cloud structures but struggle with local feature extraction and global modeling. To address these issues, we introduce VectorMamba, a novel 3D point cloud analysis network. VectorMamba employs a Vector-oriented Set Abstraction (VSA) method that integrates scalar, rotation, and scaling information into vector representations, enhancing local feature representation. Additionally, the Flash Residual MLP (FlaResMLP) module improves generalization and efficiency by leveraging anisotropic functions and explicit positional embeddings. To address global modeling challenges, we propose the PosMamba Block, a state-space-based module that incorporates positional encoding to preserve spatial information and mitigate the loss of geometric context in deeper layers. Experimental results on the ModelNet40 classification dataset, ShapeNetPart part segmentation dataset, and S3DIS semantic segmentation dataset demonstrate that VectorMamba outperforms baseline methods and achieves competitive performance compared to other approaches. The code and dataset are openly available at <span><span>github.com/Shadow581/VectorMamba</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104255"},"PeriodicalIF":2.5,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144263742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yu Han , Weiduo Xu , Pingan Liu , Xinyu Xu , Xi Duan , Bo Qin , Xinjie Wang
{"title":"Real-time dynamic 3D geological visualization based on Octree-TEN","authors":"Yu Han , Weiduo Xu , Pingan Liu , Xinyu Xu , Xi Duan , Bo Qin , Xinjie Wang","doi":"10.1016/j.cag.2025.104259","DOIUrl":"10.1016/j.cag.2025.104259","url":null,"abstract":"<div><div>3D geological visualization offers extensive support for geographic information systems. Grid space division is a fundamental technique of 3D geological visualization. However, to our knowledge, the existing division structures give little consideration to the construction or simulation of physical processes; meanwhile, most physical systems only focus on the calculation of physical fields, failing to build indexes and control voxel units. To tackle these challenges, we propose a novel real-time dynamic 3D geological visualization method based on the Octree and Tetrahedral Network (Octree-TEN). Our method combines Octree-TEN with Position-based Dynamics (PBD) to achieve voxel-controllable PBD physical field calculations. Therefore, it is suitable for data-driven visualization of fractured-grid physical field calculations, such as landslide simulation. Furthermore, in the data pre-processing phase, i.e. the process of generating voxelized grids from raw data, we employ an enhanced Delaunay Triangulation method to improve efficiency. To build a practical visualization system, we optimize load balancing at the engine rendering stage and the Delaunay simplification stage, respectively. In the experiment, we dynamically visualize geological information containing nearly 2 million voxels, which reach 118.5 FPS on an NVIDIA GeForce 3060 GPU. It indicates that the proposed method is both effective and feasible. Moreover, our method has potential applications in other fields, such as geological disaster prediction, mineral resource exploration, and popular science education.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104259"},"PeriodicalIF":2.5,"publicationDate":"2025-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144270340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Mural image restoration with spatial geometric perception and progressive context refinement","authors":"Yumeng Zhou, Min Guo, Miao Ma","doi":"10.1016/j.cag.2025.104266","DOIUrl":"10.1016/j.cag.2025.104266","url":null,"abstract":"<div><div>Ancient murals, as invaluable cultural heritage, have long been a focal point and significant challenge in the field of cultural heritage preservation. Traditional restoration methods typically address texture and structural features separately, leading to inconsistencies between local details and the overall structure. This approach is insufficient to meet the complex demands of texture and structural restoration for ancient murals. To address this issue, this paper proposes a collaborative encoder–decoder architecture (MIR-SGPR) that achieves simultaneous restoration of texture and structural features in ancient mural images. The generator extracts shallow texture features and deep structural features through the encoder and, in conjunction with the Spatial Geometric Awareness (SGA) module, achieves precise modeling of the spatial location and directional information of damaged areas. To resolve the imbalance between local details and global semantics, this paper introduces the Progressive Contextual Refinement (PCR) network, which progressively optimizes multi-scale features and effectively integrates texture and structural information, thereby enhancing the collaborative modeling capability of local details and global structure. Furthermore, this paper proposes the Mask Reverse-Focus Mechanism (MRF), which leverages mask information to eliminate feature interference from undamaged areas, significantly improving the efficiency and accuracy of restoration. Ultimately, the generated images are optimized through both the global and local discriminators. Experimental results demonstrate that this method significantly outperforms existing state-of-the-art approaches across multiple evaluation metrics. The generated restored images exhibit superior visual consistency, detail authenticity, and overall structural recovery, providing an efficient and reliable solution for the digital preservation of ancient murals.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104266"},"PeriodicalIF":2.5,"publicationDate":"2025-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144280914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Janine Zöllner , Bernhard Preim , Jan-Willem Vahlbruch , Vivien Pottgießer , Patrick Saalfeld
{"title":"Exploration of interactive nuclide chart visualisations in virtual reality for physics education","authors":"Janine Zöllner , Bernhard Preim , Jan-Willem Vahlbruch , Vivien Pottgießer , Patrick Saalfeld","doi":"10.1016/j.cag.2025.104258","DOIUrl":"10.1016/j.cag.2025.104258","url":null,"abstract":"<div><div>Immersive virtual reality (VR) is used for various types of learning content. One fundamental but challenging part of VR applications are suitable interaction techniques. In this work, we use the example of interactive nuclide charts to investigate interaction techniques in VR. For this purpose, four variants of visualising an interactive nuclide chart for decay rows in a VR environment were implemented: The floor-freehand variant offers the possibility to move freely on the chart, the floor-controller variant enables teleportation to nuclides using the controller, the wall-freehand variant uses hand gestures to select nuclides, and the wall-controller variant also uses the controller to select nuclides on the wall. Our user study with 24 participants indicated that the wall-controller variant was favoured in terms of usability and user experience.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104258"},"PeriodicalIF":2.5,"publicationDate":"2025-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144243038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"SR-CurvANN: Advancing 3D surface reconstruction through curvature-aware neural networks","authors":"Marina Hernández-Bautista , Francisco J. Melero","doi":"10.1016/j.cag.2025.104260","DOIUrl":"10.1016/j.cag.2025.104260","url":null,"abstract":"<div><div>Incomplete or missing data in three-dimensional (3D) models can lead to erroneous or flawed renderings, limiting their usefulness in applications such as visualization, geometric computation, and 3D printing. Conventional surface-repair techniques often fail to infer complex geometric details in missing areas. Neural networks successfully address hole-filling tasks in 2D images using inpainting techniques. The combination of surface reconstruction algorithms, guided by the model’s curvature properties and the creativity of neural networks in the inpainting processes, should provide realistic results in the hole completion task. In this paper, we propose a novel method entitled SR-CurvANN (Surface Reconstruction Based on Curvature-Aware Neural Networks) that incorporates neural network-based 2D inpainting to effectively reconstruct 3D surfaces. We train the neural networks with images that represent planar representations of the curvature at vertices of hundreds of 3D models. Once the missing areas have been inferred, a coarse-to-fine surface deformation process ensures that the surface fits the reconstructed curvature image. Our proposal makes it possible to learn and generalize patterns from a wide variety of training 3D models, generating comprehensive inpainted curvature images and surfaces. Experiments conducted on 959 models with several holes have demonstrated that SR-CurvANN excels in the shape completion process, filling holes with a remarkable level of realism and precision.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104260"},"PeriodicalIF":2.5,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144223286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"IMPAVID: Enhancing incident management process compliance assessment with visual analytics","authors":"Alessandro Palma , Marco Angelini","doi":"10.1016/j.cag.2025.104243","DOIUrl":"10.1016/j.cag.2025.104243","url":null,"abstract":"<div><div>The Incident Management Process (IMP) is crucial to prevent, protect against, and respond to security incidents that impact an organization. To ensure readiness for potential alerts, the IMP must comply with security standards, which provide guidelines for managing such incidents, and organizations are expected to adhere to these standards to establish a secure-by-design approach. Evaluating an organization’s compliance with security standards is often labor-intensive, as traditional methods rely heavily on manual analysis. Incorporating automated approaches to aid decision-making presents additional challenges, such as data interpretation and correlation. To address these challenges, we present IMPAVID, a visual analytics solution designed to support the assessment of IMP compliance through process-centric techniques. IMPAVID aims to enhance the security assessor’s awareness, enabling them to make informed decisions about improving the IMP alignment with regulatory and technical standards. To ensure the context-awareness of these techniques, IMPAVID leverages a deviations taxonomy and a cost model to propose a more fine-grained analysis linking together process and technical data while allowing to focus on general root causes for non-compliance. In the literature, cost models often rely on parametric cost functions that provide a valuable solution for fine-grained assessments while introducing additional challenges related to the effort necessary for security assessors to determine suitable parameter configurations. Thus, the IMPAVID system implements additional requirements and a visual environment to support data-driven, assisted, and interactive parameter configuration during IMP compliance assessment. We validate our system by presenting a comprehensive case study based on a publicly available dataset, which includes real IMP log data from an IT company. It shows the system’s capabilities to perform IMP compliance assessment while dynamically configuring the parameters of the proposed compliance cost model, enabling more effective and efficient analysis.</div></div>","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104243"},"PeriodicalIF":2.5,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144243037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Foreword to the special section on SIBGRAPI 2023 tutorials","authors":"Rafael Piccin Torchelsen , João Paulo Lima","doi":"10.1016/j.cag.2025.104257","DOIUrl":"10.1016/j.cag.2025.104257","url":null,"abstract":"","PeriodicalId":50628,"journal":{"name":"Computers & Graphics-Uk","volume":"130 ","pages":"Article 104257"},"PeriodicalIF":2.5,"publicationDate":"2025-06-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144212902","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}