Eurographics Digital Library
Not a member yet
17121 research outputs found
Sort by
Supervised Models to Support Investigations of Ancient Coins
No full textThis paper presents the initial findings of the ongoing MML-ARCH project, which uses machine learning (ML) algorithms to create predictive, supervised models for analyzing archaeological, numismatic and physicochemical data. Specifically, the study proposes using convolutional neural network (CNN) algorithms to predict the minting year of ancient Roman Republican coins based on the iconography on the obverse and reverse.Digital HeritageDigital Technologies for CHANGES (CHANGES SESSION) - Part
InterFaceRays: Interaction-Oriented Furniture Surface Representation for Human Pose Retargeting
No full textMotion retargeting is a well-established technique in computer animation that adapts source motion to fit characters with different sizes, morphologies, or environments. Recent deep learning methods have shown promising results in retargeting character motion. However, retargeting human-object interactions to new environments, especially when furniture shapes differ significantly, remains a challenging problem. In this work, we propose a novel retargeting framework to address this challenge by combining motion generative models with optimization-based pose adaptation. Our framework operates in two stages: first, a key pose generator generates the pose of key joints that preserves the interaction state relative to the new furniture; second, final whole-body pose is determined by accommodating the key joints' poses through optimization. A crucial step in our framework is generating key poses that maintain the interaction state of the source motion. To achieve this, we introduce the Interaction Intensity Weight (IIW) and structural rays, called InterFaceRays, which together capture the interaction intensity between body parts and furniture surfaces. The IIW generator, a trained MoE-based decoder from the conditional variational autoencoder (cVAE) model, infers IIWs for the target furniture based on the source motion's interaction state. Extensive experiments demonstrate that our framework effectively retargets continuous character motion across diverse furniture configurations, with the IIW generator significantly enhancing key pose consistency. This hybrid approach offers a robust solution for motion retargeting across dissimilar furniture environments.Computer Graphics ForumRigged for Success: Character Animation and Retargeting44
DeepSwitch - A Web-based Tool for the Introduction to Visual Analysis of Spatiotemporal Processes in Oceanographic Data
No full textOceanographic data comprises a multitude of information, such as concentrations of materials, temperature, or the movement of water over time. The interpretation and analysis of such data is useful for a large number of domains which all contribute to the understanding of dynamic environmental processes. Complex visual analysis applications are typically handcrafted for the needs of domain scientists and are thus powerful tools for experts, but often require a steep learning curve and involved setups. Visualization used for education and science communication on the other hand is often limited to simplistic 2D views, where temporal aspects are mainly communicated via flip-book animation or side-by-side comparisons, thus limiting the interactive exploration of data and ease of connecting temporal and spatial features. We present DeepSwitch - a web-based visual analysis tool tailored towards an accessible exploration of dynamic processes in oceanographic data. Its design is based on requirements that make it applicable to classroom environments and its central paradigm introduces a fast switch between fixed-time and fixed-depth representation of the data, aiming to minimize cognitive load while providing versatile exploration options for dynamic processes. We demonstrate its usefulness in two analysis scenarios and present preliminary feedback from teachers.Workshop on Visualisation in Environmental Sciences (EnvirVis)Session
Next Generation 3D Face Models
No full textData driven 3D face models are an important tool for applications like facial animation, face reconstruction and tracking and can serve as a powerful prior for the complex nonrigid deformation of human faces. While linear 3D morphable models or 3DMMs have been traditionally employed by artists to cater to these applications, in the last few years several deep face models have been introduced that make use of neural networks to manipulate face shapes and offer greater flexibility while also retaining the intuitive control of traditional face models. This recent class of semantic deep face models have the potential to simplify existing facial animation workflows and enable artists to make a wider range of creative choices. However, as these neural tools are still very recent and fresh out of academic research, there is a need to start a conversation with artists and industry professionals on how such neural networks can be incorporated into existing workflows. This course aims to take a first step in this direction by providing a gentle introduction to several types of deep face models introduced in recent years by the academia and how each of them resolve several problems encountered in conventional facial animation. The primary intention of the course is to provide artists and industry professionals with an understanding of the state of art in neural 3D face models, and to inspire them to consider how these new tools can be incorporated into existing industry workflows to produce better content faster. The course will also serve the purpose of providing a gentle introduction to face modeling and animation to students looking to get familiar with the field. Experienced participants with a strong background in the field would also be able to identify possible directions for future research. The course will be presented in a lecture format with slides. Concepts from related papers will be explained in enough detail to help the audience make informed decisions on using these tools and understand their current shortcomings.Eurographics 2025 - TutorialsTutorial
REED-VAE: RE-Encode Decode Training for Iterative Image Editing with Diffusion Models
No full textWhile latent diffusion models achieve impressive image editing results, their application to iterative editing of the same image is severely restricted. When trying to apply consecutive edit operations using current models, they accumulate artifacts and noise due to repeated transitions between pixel and latent spaces. Some methods have attempted to address this limitation by performing the entire edit chain within the latent space, sacrificing flexibility by supporting only a limited, predetermined set of diffusion editing operations. We present a re-encode decode (REED) training scheme for variational autoencoders (VAEs), which promotes image quality preservation even after many iterations. Our work enables multi-method iterative image editing: users can perform a variety of iterative edit operations, with each operation building on the output of the previous one using both diffusion based operations and conventional editing techniques. We demonstrate the advantage of REED-VAE across a range of image editing scenarios, including text-based and mask-based editing frameworks. In addition, we show how REEDVAE enhances the overall editability of images, increasing the likelihood of successful and precise edit operations. We hope that this work will serve as a benchmark for the newly introduced task of multi-method image editing.Computer Graphics ForumThe Artful Edit: Stylization and Editing for Images and Video44
Image Reflection Separation via Adaptive Residual Correction and Feature Interaction Enhancement
No full textGlass reflection superimposes images from both sides of the glass, resulting in severe image quality degradation that significantly impairs the performance of downstream tasks, such as object detection and image understanding. Therefore, it is essential to separate the transmission and reflection layers. However, due to lighting conditions and the material properties of glass, the relationship between the reflected and transmitted components often involves complex linear interactions, which limit the effectiveness of existing methods. Inspired by the observation that transmission components often dominate images with reflection in real-world scenes, we propose an image reflection separation method that integrates adaptive residual correction with feature interaction enhancement. Building upon a linear combination model enhanced with residual correction, we generalize the residual term based on the physical principles of light reflection and transmission. In order to ensure precise spatial alignment between the transparent and real images, We design an image registration mechanism and propose an Adaptive Hybrid Residual Loss, which significantly enhances the model's ability to perceive differences between the transmission and reflection layers, effectively balancing the complexity of linear mixture modeling with the diversity of real-world scenarios. To further highlight the interactive features between reflection and transmission, we incorporate a cross-dimensional attention mechanism into the dual-stream architecture designed for transmission-reflection processing. Extensive experiments and ablation studies show that our method achieves state-of-the-art performance on multiple real-world benchmark datasets, with an average PSNR improvement of 0.66 dB over the current best-performing model.Pacific Graphics Conference Papers, Posters, and DemosEnhancing Image
Contextualism and Music Annotation: Exploring the Role of Digital Storytelling about a Composer's Life on Music Perception
No full textIntangible cultural heritage, especially music, is challenging to interpret due to its elusive nature and reliance on context. This study investigates how knowledge of a composer's life and historical background shapes listeners' interpretations of classical music. Based on contextualism theory, which emphasizes the importance of cultural and personal context in meaning-making, the study involved 25 participants with varying levels of familiarity with music theory and the composer Nikos Skalkotas In a two-stage process, participants first annotated four musical pieces based on initial impressions. They then revisited the pieces after having engaged with a collaborative digital narrative and virtual reality (VR) experience based on archival material about the composer, presenting his life and work. The results show that annotations became more reflective and contextually informed after the storytelling experiences, indicating enhanced empathy and historical understanding. These findings suggest that immersive, narrative-driven context can enrich music appreciation and offer valuable tools for cultural heritage education.Digital HeritageStorytelling and Interpretation in Digital Heritag
Differentiable Rendering based Part-Aware Occlusion Proxy Generation
No full textSoftware occlusion culling has become a prevalent method in modern game engines. It can significantly reduce the rendering cost by using an approximate coarse mesh (occluder) to cull hidden objects. An ideal occluder should use as few faces as possible to represent the original mesh with high culling accuracy. In contrary to mesh simplification, the process of generating a high quality occlusion proxy is not well-established. Existing methods, which simply treat the mesh as a single entity, fall short in addressing complex models with interior structures. By leveraging advanced neural segmentation techniques and the optimization capabilities of differentiable rendering, in combination with a thoughtfully designed part-aware shape fitting and camera placement strategy, our approach can generate high-quality occlusion proxy mesh applicable across a diverse range of models with satisfactory precision, recall and very few faces. Moreover, extensive experiments compellingly demonstrate that our method substantially outperforms both state-of-the-art methodologies and commercial tools in terms of occlusion quality and effectiveness.Computer Graphics ForumThe Shape of Rendering44
Navigating High-Dimensional Backstage: A Guide for Exploring Literature for the Reliable Use of Dimensionality Reduction
No full textVisual analytics using dimensionality reduction (DR) can easily be unreliable for various reasons, e.g., inherent distortions in representing the original data. The literature has thus proposed a wide range of methodologies to make DR-based visual analytics reliable. However, the diversity and extensiveness of the literature can leave novice analysts and researchers uncertain about where to begin and proceed. To address this problem, we propose a guide for reading papers for reliable visual analytics with DR. Relying on the previous classification of the relevant literature, our guide helps both practitioners to (1) assess their current DR expertise and (2) identify papers that will further enhance their understanding. Interview studies with three experts in DR and data visualizations validate the significance, comprehensiveness, and usefulness of our guide.EuroVis 2025 - Short PapersSystems and Application
Future Challenges and Unsolved Problems in Health Visualization
No full textWith the growing popularity of wearable devices, health-related sensors, electronic health records (EHR), population health records (PopHR), computational biology and simulation, imaging data such as CT and MRI scans, and X-rays, the volume of digital health data is growing rapidly. Large volumes of heterogeneous health data require advanced visualization and visual analytics systems to uncover valuable insight buried in complex sources of data. As a rapidly evolving sub-field of visualization and visual analytics, many interactive health visualization systems have been proposed, developed, and evaluated by clinicians to support effective clinical analysis and decision making. Despite the growing progress in the field, many challenges and unsolved problems remain. The health-related problems that we face today are a clear sign of the growing need to progress in this area. This panel presents an open discussion of the top future challenges and unsolved problems in health and healthrelated visualization. The panel features experts with a range of different backgrounds covering a variety of health-related perspectives. This panel provides a valuable overview of health-related visualization revealing both mature areas and future research directions.EuroVis 2025 - Panels and TutorialsPanel