Eurographics Digital Library
Not a member yet
17121 research outputs found
Sort by
PGV 2025: Frontmatter
No full textEurographics Symposium on Parallel Graphics and VisualizationEurographics Symposium on Parallel Graphics and Visualization 202
Axis-Normalized Ray-Box Intersection
No full textRay-axis aligned bounding box intersection tests play a crucial role in the runtime performance of many rendering applications, driven not by complexity but mainly by the volume of tests required. While existing solutions were believed to be pretty much optimal in terms of runtime on current hardware, our paper introduces a new intersection test requiring fewer arithmetic operations compared to all previous methods. By transforming the ray we eliminate the need for one third of the traditional bounding-slab tests and achieve a speed enhancement of approximately 13.8% or 10.9%, depending on the compiler.We present detailed runtime analyses in various scenarios.Computer Graphics ForumReal-Time Rendering: Fast, Furious, and Accurate44
Engaging History Through Play: The Potential of Digitally-Enhanced Tabletop Role Playing Games for Promoting Historical Empathy in Museums
No full textIn an era of rapidly advancing digital technologies and changing visitor expectations, museums face the ongoing challenge of making history engaging and accessible to diverse audiences, through encouraging active participation and emotional connections. In this study, we investigate the potential of digitally enhanced tabletop role-playing games (TTRPGs), incorporating digital technologies such as virtual reality and other interactive media, in fostering historical empathy and learning within the museum setting. Thirteen museum educators with diverse professional backgrounds provided key insights into the objectives and challenges of museum educational programs and the potential use of role-playing activities in them. Results reveal their main challenge is balancing creativity, interactivity and emotional impact with practical constraints such as space, time, and limited resources. While digitally enhanced TTRPGs are seen as an effective tool for enabling active engagement with historical figures and events in immersive, participatory ways, there is a need for clear guidelines and adaptability in their design, to ensure alignment with both the goals of the museum and the diverse needs of history learners.Digital HeritageExplorative Approaches for Histor
Digital Technologies for the Conservation and Enhancement of the Castle of Gaeta: An Inclusive and Participatory Approach
No full textThe Angevin Castle of Gaeta, located in southern Lazio, represents a unique architectural complex and preserves a rich cultural heritage, both tangible and intangible. With a history spanning from the 13th century to the present day, the Castle has long served as a vital point of reference for the local community. State-owned, since 1995 it has been granted in ''free and perpetual use'' to the University of Cassino and Southern Lazio, which employs it as a venue for higher education, seminars, conferences, and cultural initiatives. These initiatives are promoted and coordinated by the Rector's Delegate for the Dissemination of Culture and Knowledge, Professor Ivana Bruno. This paper outlines the objectives and operational methods of the project ''The Castle of Gaeta: Virtual Technologies and New Communicative Languages of Public History for the Narration and Musealization of Memory,'' developed within the framework of the program CHANGES - Cultural Heritage Active Innovation for Sustainable Society. This initiative involves a partnership of 11 universities, 4 research institutions, 3 advanced studies schools, 6 enterprises, and the Center of Excellence of the Technological District for Cultural Heritage and Activities in the Lazio Region (DTC Lazio). Aiming to safeguard, enhance, and disseminate knowledge of the monument's tangible and intangible heritage, the project adopts an innovative approach centered on accessibility and broad public engagement. This is achieved through the use of advanced digital technologies and new communicative methods designed to narrate memory within its original historical context.Digital HeritagePoster
Benchmarking Visual Language Models on Standardized Visualization Literacy Tests
No full textThe increasing integration of Visual Language Models (VLMs) into visualization systems demands a comprehensive understanding of their visual interpretation capabilities and constraints. While existing research has examined individual models, systematic comparisons of VLMs' visualization literacy remain unexplored. We bridge this gap through a rigorous, first-ofits- kind evaluation of four leading VLMs (GPT-4, Claude, Gemini, and Llama) using standardized assessments: the Visualization Literacy Assessment Test (VLAT) and Critical Thinking Assessment for Literacy in Visualizations (CALVI). Our methodology uniquely combines randomized trials with structured prompting techniques to control for order effects and response variability - a critical consideration overlooked in many VLM evaluations. Our analysis reveals that while specific models demonstrate competence in basic chart interpretation (Claude achieving 67.9% accuracy on VLAT), all models exhibit substantial difficulties in identifying misleading visualization elements (maximum 30.0% accuracy on CALVI). We uncover distinct performance patterns: strong capabilities in interpreting conventional charts like line charts (76-96% accuracy) and detecting hierarchical structures (80-100% accuracy), but consistent difficulties with data-dense visualizations involving multiple encodings (bubble charts: 18.6-61.4%) and anomaly detection (25-30% accuracy). Significantly, we observe distinct uncertainty management behavior across models, with Gemini displaying heightened caution (22.5% question omission) compared to others (7-8%). These findings provide crucial insights for the visualization community by establishing reliable VLM evaluation benchmarks, identifying areas where current models fall short, and highlighting the need for targeted improvements in VLM architectures for visualization tasks. To promote reproducibility, encourage further research, and facilitate benchmarking of future VLMs, our complete evaluation framework, including code, prompts, and analysis scripts, is available at https://github.com/washuvis/VisLit-VLM-Eval.Computer Graphics ForumAI-Enhanced Visualizatio
Opportunities and Tools from the Cultural Heritage Cloud
No full textThis four-hour workshop will introduce participants to the European Collaborative Cloud for Cultural Heritage (ECCCH), also called Cultural Heritage Cloud, an initiative aimed at fostering collaboration, innovation, and digital preservation within the cultural heritage sector. The Cloud is being set up by the ECHOES project, and other projects of the ECCCH family (i.e. projects that are funded by the same family of EU calls) will contribute to it. The three first projects AUTOMATA, TEXTaiLES, HERITALISE, that are currently funded, will also participate in the proposed workshop. Through a combination of presentations and group discussions, attendees will discover how the Cultural Heritage Cloud and the associated applications, brought in by ECHOES and the 3 sister projects, will be able to enhance their research projects and contribute to the broader cultural heritage community. By the end of the workshop, participants will have a clear understanding of how to utilize the Cultural Heritage Cloud to foster sustainable collaboration and address challenges in the cultural heritage domain.Digital HeritageWorkshop
Gaussian Splatting for Large-Scale Aerial Scene Reconstruction From Ultra-High-Resolution Images
No full textUsing 3D Gaussian splatting to reconstruct large-scale aerial scenes from ultra-high-resolution images is still a challenge problem because of two memory bottlenecks - excessive Gaussian primitives and the tensor sizes for ultra-high-resolution images. In this paper, we propose a task partitioning algorithm that operates in both object and image space to generate a set of small-scale subtasks. Each subtask's memory footprints is strictly limited, enabling training on a single high-end consumer-grade GPU. More specifically, Gaussian primitives are clustered into blocks in object space, and the input images are partitioned into sub-images according to the projected footprints of these blocks. This dual-space partitioning significantly reduces training memory requirements. During subtask training, we propose a depth comparison method to generate a mask map for each sub-image. This mask map isolates pixels primarily contributed by the Gaussian primitives of the current subtask, excluding all other pixels from training. Experimental results demonstrate that our method successfully achieves large-scale aerial scene reconstruction using 9K resolution images on a single RTX 4090 GPU. The novel views synthesized by our method retain significantly more details than those from current state-of-the-art methods.Computer Graphics ForumGaussian Splatting44
Gridded Visualization of Statistical Trees for High-Dimensional Multipartite Data in Systems Genetics
No full textIn systems genetics and other multi-omics research, exploring high-dimensional relationships among molecular and physiological variables across individuals poses significant challenges. We present the Gridded Trees interface, a novel interactive visualization tool designed to facilitate the exploration of conditional inference trees, which are hierarchical models of relationships in these complex datasets. Traditional static tools struggle to reveal patterns in tree-structured data, but the Gridded Trees interface provides interactive, coordinated views, allowing users to navigate between overview and detail, filter data dynamically, and compare molecular-physiological relationships across subgroups. By combining filtering techniques, strip plots, Sankey diagrams, and small multiples, the Gridded Trees interface enhances exploratory data analysis and supports hypothesis generation. In our systems genetics research use case, this tool has revealed significant associations among microbial populations and addiction-related behavioral traits in genetically diverse mice. The Gridded Trees interface suggests broad potential for visualizing hierarchical and multipartite data across domains. A preprint of this paper as well as Supplemental Materials are available on OSF at https://osf.io/9emn5/.Computer Graphics ForumHonourable Mentio
Interactive Particle System with Integrated Marching Cubes Algorithm
No full textThis paper highlights a web-based platform that merges real-time particle simulations with Marching Cubes mesh generation, enabling fluid-like shapes to emerge from a dynamic emitter. The project emphasizes a modular architecture, ensuring flexible data flow and performance. A web-based development emphasizes accessibility and ease of use, differentiating it from traditional desktop-based visualization tools by enabling intuitive, immediate interaction directly within the browser. Key accomplishments include real-time parameter manipulation, immediate visual feedback and careful handling of computational complexity. Lessons learned range from optimizing efficiency under heavy particle loads to refining user experience.Eurographics 2025 - Education PapersEducation
OCCAM: Occlusion-aware Completeness via Coverage Analysis with Monte Carlo Sampling
No full textOcclusion is a common challenge in indoor 3D scanning, often leading to incomplete scene reconstructions due to unobserved surfaces. Estimating it during the scan process can improve the quality of the acquired scenes tremendously. However, most existing methods for estimating scan completeness require access to ground-truth data, an assumption that rarely holds in practical settings. We introduce OCCAM (Occlusion-aware Completeness via Coverage Analysis with Monte Carlo sampling), a lightweight method that estimates global scan coverage without requiring surface reconstruction. It casts randomized rays from within the scanned volume to identify visibility gaps, without relying on mesh connectivity or external reference geometry. In contrast to occupancy grid mapping methods, which model local space coverage from the scanner's perspective, OCCAM evaluates broader scene visibility to detect whether large surface regions remain unscanned. Experimental results on synthetic and real-world benchmark datasets show that the proposed method is fast to compute (processing 100K-point scans in under one second), simple to implement, and produces a compact signal that supports both coverage assessment and scan guidance in indoor environments.Computer Graphics and Visual Computing (CGVC)Computer Vision for Graphic