Eurographics Digital Library
Not a member yet
17121 research outputs found
Sort by
StyleBlend: Enhancing Style-Specific Content Creation in Text-to-Image Diffusion Models
No full textSynthesizing visually impressive images that seamlessly align both text prompts and specific artistic styles remains a significant challenge in Text-to-Image (T2I) diffusion models. This paper introduces StyleBlend, a method designed to learn and apply style representations from a limited set of reference images, enabling content synthesis of both text-aligned and stylistically coherent. Our approach uniquely decomposes style into two components, composition and texture, each learned through different strategies. We then leverage two synthesis branches, each focusing on a corresponding style component, to facilitate effective style blending through shared features without affecting content generation. StyleBlend addresses the common issues of text misalignment and weak style representation that previous methods have struggled with. Extensive qualitative and quantitative comparisons demonstrate the superiority of our approach.Computer Graphics ForumThe Artful Edit: Stylization and Editing for Images and Video44
Spatio-Temporal Dithering for Order-Independent Transparency on Ray Tracing Hardware
No full textEfficient rendering of many transparent surfaces is a challenging problem in real-time ray tracing. We introduce an alternative approach to conventional order-independent transparency (OIT) techniques: our method interprets the alpha channel as coverage and uses state-of-the-art temporal anti-aliasing techniques to accumulate transparency over multiple frames. By efficiently utilizing ray tracing hardware and its early ray termination capabilities, our method reduces computational costs compared to conventional OIT methods. Furthermore, our approach shades only one fragment per pixel, significantly lowering the shading workload and improving frame rate stability. Despite relying on temporal accumulation, our technique performs well in dynamic scenes.Eurographics Symposium on RenderingReal-Time Renderin
Implicit Field-Based Stylization of 2D and 3D Liquid Animations
No full textCombining physically-based simulation with stylized visual effects not only requires to change the aspect of surfaces, but their shapes as well. We describe a new expressive rendering method for liquid animations, which can be used on top of any preexisting particle-based simulation. Our solution builds on visual particles that carry both water and air distributions, both evolving through particle history based on kinematics information from the simulation. These density fields are combined at each frame to create the implicit iso-surface of interest, rendered in adapted style. By defining series of visual particle states, we parametrize this model to capture the typical stylized geometry of water bodies used to highlight dynamic motion in paintings and cartoons, such as elongating droplets, concavities carved at the crest of breaking waves, and stylized air-water mixtures such as bubbles and foam. Regardless of the 2D or 3D nature of the input simulation, our solution maintains temporal coherence and ensures that water bodies keep an approximately constant surface in 2D, resp. or volume in 3D, over time. Finally, we conducted a user study to show the effectiveness of our method against state of the art AI-based tools, in a variety of animation scenarios where stylized shapes are needed.Smart Tools and Applications in Graphics - Eurographics Italian Chapter ConferenceGeometry Processin
BlendSim: Simulation on Parametric Blendshapes using Spacetime Projective Dynamics
No full textWe propose BlendSim, a novel framework for editable simulation using spacetime optimization on the lightweight animation representation. Traditional spacetime control methods suffer from a high computational complexity, which limits their use in interactive animation. The proposed approach effectively reduces the dimensionality of the problem by representing the motion trajectories of each vertex using continuous parametric Bézier splines with variable keyframe times. Because this mesh animation representation is continuous and fully differentiable, it can be optimized such that it follows the laws of physics under various constraints. The proposed method also integrates constraints, such as collisions and cyclic motion, making it suitable for real-world applications where seamless looping and physical interactions are required. Leveraging projective dynamics, we further enhance the computational efficiency by decoupling the optimization into local parallelizable and global quadratic steps, enabling a fast and stable simulation. In addition, BlendSim is compatible with modern animation workflows and file formats, such as the glTF, making it practical way for authoring and transferring mesh animation.Computer Graphics ForumSoft Bodies, Strands, and Silks44
Engaging Young Minds: Learnings from Data Visualization Workshops with Children and Teenagers
No full textAs the role of visual data representations in everyday information consumption is increasing, the importance of teaching data visualization literacy to lay audiences, such as children and teenagers, has grown. Research involving children and teenagers poses unique challenges, resulting in limited literature on teaching data visualization to this audience. This paper reports on the design and implementation of five educational workshops for children and teenagers aged 7 to 16, conducted over three years and teaching data visualization in the context of themes such as climate change, detective work, and social media. The workshops emphasized interactive learning and hands-on projects to engage participants and enhance their understanding of data visualization concepts and critical thinking skills. Key observations include the importance of tailoring content to developmental stages, using storytelling and gamification to sustain interest, and fostering collaboration through activities in small groups. By reflecting on these experiences, the paper contributes to the growing body of research on visualization education and offers practical resources for educators seeking to make data literacy more accessible to younger audiences.EuroVis 2025 - Education PapersEducation Papers Session
The RePAIR Project: Datasets for archaeological and restoration studies in Pompeii
No full textThe work will focus on the RePAIR project, an acronym that means Reconstructing the Past: Artificial Intelligence and Robotics meets Cultural Heritage. The case study is the Archaeological Park of Pompeii in which the fragments of wall paintings from two buildings buried by the volcanic eruption of AD 79, the House of Painters at Work and of the Schola Armaturarum have been chosen as the focus of the project. We concentrated on developing an innovative technology to be used in the physical reconstruction of archaeological artefacts with the help of artificial intelligence and a robotic infrastructure. Two Datasets are used to allow the artificial intelligence to recognise the 3D volumes of each fragment and the iconography of the pompeian wall paintings.Digital HeritagePredictive Analysis, AI, Simulation, and Novel Computational Method
Designing Personalized Cultural Heritage Serious Games through Gamification, AI and Augmented Reality
No full textDigital technologies are transforming how cultural heritage is accessed, experienced, and preserved, but many applications remain static, lacking personalization and adaptivity. This paper introduces a modular reference architecture that underpins a serious game-based platform integrating Artificial Intelligence (AI), Augmented Reality (AR), and gamification to support personalized learning journeys in cultural heritage contexts. The system integrates adaptive learning pathways, meaningful gamification, and augmented reality to support user-centric engagement. AI-driven recommendations personalize the path of each visitor through path-based learning activities, adapting to individual behavior, preferences, and cognitive profiles. A conversational AI assistant and spatial AR overlays enhance narrative immersion and support situated learning. The architecture enables curators to create flexible, pedagogically informed content through an intuitive visual design interface. A preliminary prototype demonstrates the technical feasibility of the system and how the integration of AI, gamification, and augmented reality can support the development of inclusive, engaging, and educational cultural experiences for diverse audiences.Digital HeritageModern Technologies for Serious Gaming in Cultural Heritag
Drone-Based Magnetic Survey: Testing a New Approach for Archaeological Prospection
No full textIn recent times, drones equipped with a variety of sensors, such as multispectral, thermal, and LiDAR, have been increasingly used in various archaeological tasks. There is also a growing interest in using geophysical sensors, like magnetometry (MAG), on Unmanned Aerial Vehicles (UAVs). One of the key challenges is selecting the most suitable and effective systems for archaeological purposes. Succinctly, this paper aims to assess the inner workflow established of Drone-based MAG system and briefly highlight its advantages and limitations in the archaeology field. Instead of discussing specific market-available systems, this essay provides a comprehensive overview of Drone-based MAG applications.Digital HeritageDigitization Tools and Application
Neural Resampling with Optimized Candidate Allocation
No full textWe propose a novel framework that accelerates Monte Carlo rendering with the help of machine learning. Unlike previous works that learn parametric distributions that can be sampled directly, our method learns the 5-dimensional unnormalized incident radiance field and samples its product with the material response (BRDF) through Resampled Importance Sampling. This allows for more flexible network architectures that can be used to improve upon existing path guiding approaches and can also be reused for other tasks such as radiance caching. To reduce the cost of resampling, we derive optimized spatially-varying candidate counts to maximize the efficiency of the render process. We designed our method to accelerate CPU production renders by benefiting from otherwise idle GPU resources without need of intrusive changes to the renderer. We compare our approach against state-of-the-art path guiding methods, both neural and non-neural, and demonstrate significant variance reduction at equal render times on production scenes.Eurographics Symposium on RenderingSampling and Guidin
PALEOTWIN. A Platform for Creating Integrated Digital Experiences: the case study of the Collezione di Geologia ''Museo Giovanni Capellini''
No full textThe ''PALEOTWIN'' Project aims to develop a digital platform focused on the creation of a central repository for managing and retrieving digital materials related to museum cultural heritage. Here we present a case study applied to the Collezione di Geologia ''Museo Giovanni Capellini'', part of the Bologna Museum University Network. This centralized repository will support various digital consumption formats (texts, audio, 3D reconstructions, augmented reality, and virtual reality reconstructions), including customizable pathways based on semantic logic, to enhance accessibility and usability, particularly for users with disabilities. The project includes the implementation of a ''gamified'' mode and immersive functionalities for an augmented reality (AR) and virtual reality (VR) experience, enabling advanced interactions with digital models and 3D environments. Here we present the system architecture and the design of the semantic repository which represent the backbone of the project. The entire proposed architecture is designed to be easily configurable and expandable, thanks to a software and hardware framework optimized for user interaction and ease of use.Digital HeritageDigital Technologies for CHANGES (CHANGES SESSION) - Part