45-Issue 2

Permanent URI for this collection

EG 2026
Animating Humans with Gestures and Style
Conversational Gesture Model (CGM): Extending Speaker-Centric Audio-Driven Motion Generation to Full Conversation Gestures
Koren Tomer, Rosenthal Adi, Friedman Doron, and Shamir Ariel
Skeletal-Driven Animation of Anatomical Humans via Neural Deformation Gradients
Nolte Gerrit, Kemper Fabian, Schwanecke Ulrich, and Botsch Mario
Dance Like a Chicken: Low-Rank Stylization for Human Motion Diffusion
Haim Sawdayee, Chuan Guo, Guy Tevet, Bing Zhou, Jian Wang, and Amit Haim Bermano
SkinCells: Sparse Skinning using Voronoi Cells
Larionov Egor, Santesteban Igor, Chen Hsiao-yu, Lin Gene Wei-Chin, Herholz Philipp, Goldade Ryan, Kavan Ladislav, Roble Doug, and Stuyck Tuur
Motion in the Wild: From Individuals to Crowds
VQ-Style: Disentangling Style and Content in Motion with Residual Quantized Representations
Fatemeh Zargarbashi, Dhruv Agrawal, Jakob Buhmann, Martin Guay, Stelian Coros, and Robert W. Sumner
Physics-Based Motion Tracking of Contact-Rich Interacting Characters
Zhang Xiaotang, Chang Ziyi, Men Qianhui, and Shum Hubert P. H.
Step2Motion: Locomotion Reconstruction from Pressure Sensing Insoles
Ponton Jose Luis, Alvarado Eduardo, Foo Lin Geng, Nuria Pelechano, Andujar Carlos, and Habermann Marc
ContactVision: Learning Foot Contact from Video for Physically Plausible Gait Animation
Kim DaeYong, Yi Gyuseok, and Yu Ri
Digital Humans: From Capture to Control
DexterCap: Affordable and Automated Capture of Complex Hand-Object Interactions
Liang Yutong, Xu Shiyi, Zhang Yulong, Zhan Bowen, Zhang He, and Liu Libin
Improving Facial Rig Semantics for Tracking and Retargeting
Omens Dalton, Thurman Allise, Yu Jihun, and Fedkiw Ron
CANRIG: Cross-Attention Neural Face Rigging with Variable Local Control
Mohammadi Arad, Weiss Sebastian, Buhmann Jakob, Ciccone Loic, Sumner W. Robert, Bradley Derek, and Guay Martin
GTAvatar: Bridging Gaussian Splatting and Texture Mapping for Relightable and Editable Gaussian Avatars
Baert Kelian, Younes Mae, Bourel Francois, Christie Marc, and Boukhayma Adnane
Neuralocks: Real-Time Dynamic Neural Hair Simulation
Lin Gene Wei-Chin, Larionov Egor, Chen Hsiao-yu, Roble Doug, and Stuyck Tuur
2D and Beyond: Stylized Animation and Reconstruction
3D Character Reconstruction from Hand-drawn Model Sheets
Yoon Hyejeong, Jang Wonjong, Hwang Yoonha, and Lee Seungyong
Generative Cutout Animation
Puhachov Ivan, Aigerman Noam, Groueix Thibault, and Bessmeltsev Mikhail
Mixed Super-Circles
Hohnadel Emile, Métivet Thibaut, and Bertails-Descoubes Florence
Vector sketch animation generation with differentialable motion trajectories
Zhu Xinding, Yang Xinye, Zheng Shuyang, Zhang Zhexin, Gao Fei, Huang Jing, and Chen Jiazhou
From Pixels to Scenes: 3D Reconstruction and Generation
ZeroScene: A Zero-Shot Framework for 3D Scene Generation from a Single Image and Controllable Texture Editing
Tang Xiang, Li Ruotong, and Fan Xiaopeng
GS-2M: Material-aware Gaussian Splatting for High-fidelity Mesh Reconstruction
Nguyen Dinh Minh, Avenhaus Malte, and Lindemeier Thomas
Layer3D: A 3D Layered Representation for Multiview Vector Graphics
Guan Zhongyue, Hu Yixin, and Wang Zeyu
GeoFusionLRM: Geometry-Aware Self-Correction for Consistent 3D Reconstruction
Yildirim Ahmet Burak, Saygin Tuna, Ceylan Duygu, and Dundar Aysegul
UniCross3D: Unified Cross-View and Cross-Domain Diffusion for Consistent Single-Image 3D Generation
U-Chae Jun, Jaeeun Ko, and Jiwoo Kang
Maps and Meshes: Parameterization and Geometry Processing
TABI: Tight and Balanced Interactive Atlas Packing
Gu Floria, Vining Nicholas, and Sheffer Alla
Volume Quantization with Flexible Singularities for Hexahedral Meshing
Brückler Hendrik and Campen Marcel
Fast Injective Mesh Parameterization via Beltrami Coefficient Prolongation
Fargion Guy and Weber Ofir
DiskScissors: Cutting Arbitrary-Topology Solids for Bijective Mapping
Hinderink Steffen, and Campen Marcel
Learning Surface and Scene Representations
Mesh Processing Non-Meshes via Neural Displacement Fields
Noma Yuta, Wang Zhecheng, Liu Chenxi, Singh Karan, and Jacobson Alec
Basis Networks: Learning basis functions for free-form triangulations
Djuren Tobias and Alexa Marc
Self-supervised Learning of Fine-to-Coarse Cuboid Shape Abstraction
Gregor Kobsik, Morten Henkel, Yanjiang He, Victor Czech, Tim Elsner, Isaak Lim, and Leif Kobbelt
TLC-Plan: A Two-Level Codebook Based Network for End-to-End Vector Floorplan Generation
Xiong Biao, Peng Zhen, Ping Wang, Liu Qiegen, and Zhong Xian
Floorplan Generation by Alternating Geometry and Semantics Optimization
Wu Wenming, Hu Sizhe, Liu Ligang, Zheng Liping, and Fu Xiao-Ming
Hierarchical Geometry: Optimization and Simplification
Convex Primitive Decomposition for Collision Detection
Knodt Julian and Gao Xifeng
Construction of clustered HLOD with As-Simplified-As-Possible boundaries
Ladeuil Mathieu, Trabucato Marc, Vaisse Alexis, and Faraj Noura
Hierarchical Optimization of the As-Rigid-As-Possible Energy
Meyer Hendrik, Bickel Bernd, and Alexa Marc
Embedding Optimization of Layouts via Distortion Minimization
Heuschling Alexandra, Lim Isaak, and Kobbelt Leif
Contouring Signed Distance Fields by Approximating Gradients
Kohlbrenner Maximilian and Alexa Marc
Parametric and Structured Geometry
CADrawer: Autoregressive CAD Generation from 3D Sketches
Li Yuanbo, Manfredi Gilda, Kriel Henro, Hao Chengye, Xu Xianghao, Bousseau Adrien, and Ritchie Daniel
Differentiable variable fonts
Parikh Kinjal, Kaufman Danny M., Levin David I.W., and Jacobson Alec
2D Piecewise Linear Scalar Fields with Invertible Integral Lines
Erxleben Timm Leon, Motejat Michael, Rössl Christian, and Theisel Holger
Register-Efficient Linear-Time Evaluation in the Bernstein Basis
Valasek Gábor and Horváth Anna Lili
Structural Geometry: From Fabrication to Fracture
Field-Aligned Surface-Filling Curve via Implicit Stitching
Cocco Giovanni and Chermain Xavier
Strain-Field Based Segmentation for Fabric Formwork
Sati Abhinit, Bao Tiffany, Tedi Jeff, Chien Edward, and Whiting Emily
Designing inflatable shells using unstructured meshes
He Siyuan, Lebée Arthur, and Skouras Mélina
Diffusion and Beyond: Controlled Image Generation and Stylization
Graph-based Black and White Stylization
Sattari Javid Ali, Lord Jimmy, and Mould David
Palette Aligned Image Diffusion
Aharoni Elad, Porat Noy, Lischinski Dani, and Shamir Ariel
Latent Diffusion-GAN: Adversarial Learning in the Autoencoded Latent Space
Jun U-Chae, Ko Jaeeun, and Kang Jiwoo
Edge-preserving noise for diffusion models
Vandersanden Jente, Holl Sascha, Huang Xingchang, and Singh Gurprit
TextFlux: An OCR-Free DiT Model for High-Fidelity Multilingual Scene Text Synthesis
Yu Xie, Jielei Zhang, Pengyu Chen, Weihang Wang, Longwen Gao, Peiyi Li, Qian Qiao, and Zhouhui Lian
Temporal Vision: Video Generation, Pose, and Narrative
Story2Board: A Training-Free Approach for Expressive Visual Storytelling
Dinkevich David, Levy Matan, Avrahami Omri, Samuel Dvir, and Lischinski Dani
SAGE: Structure-Aware Generative Video Transitions between Diverse Clips
Kan Mia, Liu Yilin, and Mitra J. Niloy
MultiCOIN: Multi-Modal COntrollable Inbetweening
Tanveer Maham, Zhou Yang, Niklaus Simon, Mahdavi Amiri Ali, Zhang Hao (Richard), Singh Krishna Kumar, and Zhao Nanxuan
SEE4D: Pose-Free 4D Generation via Auto-Regressive Video Inpainting
Dongyue Lu, Ao Liang, Tianxin Huang, Xiao Fu, Yuyang Zhao, Baorui Ma, Liang Pan, Wei Yin, Lingdong Kong, Wei Tsang Ooi, and Ziwei Liu
Enhancing Robust Category-Agnostic Pose Estimation through Multi-Modal Feature Alignment
Li Boxuan and Liu Juan
LeafFit: Plant Assets Creation from 3D Gaussian Splatting
Luo Chang and Umetani Nobuyuki
TreeON: Reconstructing 3D Tree Point Clouds from Orthophotos and Heightmaps
Grammatikaki Angeliki, Eschner Johannes, Hermosilla Pedro, Argudo Oscar, and Waldner Manuela
HeatMat: Simulation of City Material Impact on Urban Heat Island Effect
Reinbigler Marie, Rouffet Romain, Naylor Peter, Czerkawski Mikolaj, Dionelis Nikolaos, Brunet Elisabeth, Fetita Catalin, and Martin Rosalie
Authoring Terrestrial Planets with Diffusion Models
Borg Oliver, Gain James, Guérin Éric, Peytavie Adrien, Cani Marie-Paule, Galin Eric, and Cordonnier Guillaume
Terrain Synthesis and Authoring based on Iso-Contours
Huftier Benoit, Schott Hugo, Galin Eric, Argudo Oscar, Peytavie Adrien, and Guérin Eric
Structured for Speed: Spatial Representations for Real-Time Rendering
Real-Time Rendering of Dynamic Line Sets using Voxel Ray Tracing
Kraaijeveld Bram, Jalba Andrei C., Vilanova Anna, and Chamberland Maxime
EBOAT: Error-Bounded Adaptive Tessellation of Singularities for Real-Time Catmull-Clark Subdivision Surfaces Rendering
Yajun Zeng, Yang Lu, Cong Chen, Ruicheng Xiong, and Ligang Liu
NAADF: Globally Illuminated VoxelWorlds Accelerated with Nested Axis-Aligned Distance Fields
Ulschmid Annalena, Macho Jonas, Ott Marvin, Wimmer Michael, and Ohrhallinger Stefan
Covering the Surface: Texture Synthesis, Patterns, and Compression
Real-time by-example texture synthesis and filtering using local statistics exchange
Lutz Nicolas and Gilet Guillaume
Variable-Rate Texture Compression: Real-Time Rendering with JPEG
Kristmann Elias, Wimmer Michael, and Schütz Markus
ProcTex: Consistent and Interactive Text-to-texture Synthesis for Part-based Procedural Models
Xu Ruiqi, Zhu Zihan, Ahlbrand Benjamin, Sridhar Srinath, and Ritchie Daniel
Lightmap Compression with Color-Coherent UV Clustering and Cascade Texture Optimization
Chen Dehan, Huang Hongyu, Luo Yuzhe, Xu Hao, Zhang Yuqing, Yang Sipeng, Gao Xifeng, Cai Heng, Li Chao, and Jin Xiaogang
Controllable Intrinsic Surface Pattern Generation Using Slime Mold Simulations
Layton Jeffrey, Samavati Faramarz, and Runions Adam
Light Transport: Sampling, Waves, and Denoising
Wave Tracing: Generalizing The Path Integral To Wave Optics
Steinberg Shlomi and Pharr Matt
Gradient-Domain ReSTIR Path Tracing
Wang Yu-Chen, Wyman Chris, Kettunen Markus, Lin Daqi, Wu Lifan, and Zhao Shuang
Statistical Denoising of Transient Rendering
Pueyo-Ciutad Oscar, Alvaro Lopez, and Gutierrez Diego
Stochastic Pairwise MIS for Unbiased Large-Kernel Reuse in Real-Time
Trevor Hedstrom, Markus Kettunen, Daqi Lin, Chris Wyman, and Tzu-Mao Li
Deep Residual Combiner: A Learned Fusion of Spatial, Temporal, and Multiscale Correlated Pixel Estimates
Zhou Weijie, Hughes Euan, and Hachisuka Toshiya
Neural Appearance: Reflectance, Irradiance, and Light Transport
Neural Progressive Photon Mapping
Benoist Justin, Litalien Joey, and Gruson Adrien
Neural Local Inter-reflection Modeling for Garment Fold Rendering
Son Jooeun, Ryu Nuri, Kim Gyoonseo, Lee Joo Ho, and Lee Seungyong
Real-time Rendering with a Neural Irradiance Volume
Arno Coomans, Giacomo Nazzaro, Edoardo A. Dominici, Christian Döring, Floor Verhoeven, Konstantinos Vardis, and Markus Steinberger
A Real-Time Multi-Scale Neural Representation for Complex Surface Reflectance
Timonen Heikki, Kemppinen Pauli, and Lehtinen Jaakko
Measuring and Modeling Material Appearance
High-Gloss SVBRDF Capture Using Bounce Light
Iser Tomáš, Ardelean Andrei-Timotei, and Weyrich Tim
A Texture-Free Multi-Scale Model for Surface-Based Rendering of Knitted Fabrics
Khattar Apoorv, Aubry Jean-Marie, Yan Ling-Qi, and Montazeri Zahra
A Discrete Polydisperse Anisotropic BSDF Model based on the Micrograin Framework
Xu Kewei, Lucas Simon, Ribardiere Mickael, Bringier Benjamin, and Barla Pascal
HiMat: DiT-based Ultra-High Resolution SVBRDF Generation
Wang Zixiong, Yang Jian, Hu Yiwei, Hašan Miloš, and Wang Beibei
Advancing 3D Gaussian Splatting
Multi-Spectral Gaussian Splatting with Neural Color Representation
Meyer Lukas, Grün Josef, Weiherer Maximilian, Egger Bernhard, Stamminger Marc, and Franke Linus
RotGS: Rotation-Guided 3D Gaussian Splatting for Turntable Sequences without Structure-from-Motion
Kyumin Kim, Dohae Lee, Hanul Baek, and In-Kwon Lee
Adaptive Spatio-Temporal 3D Gaussian Splatting for Scenes with Oscillatory Motion
Tzathas Petros, Hu Jeffrey, Meuleman Andréas, Cordonnier Guillaume, Drettakis George
OUGS: Active View Selection via Object-aware Uncertainty Estimation in 3DGS
Haiyi Li, Qi Chen, Denis Kalkofen, and Hsiang-Ting Chen
Splat-based Metal Artifact Reduction in Cone-Beam CT via Polychromatic Modeling
Choi Kiseok, Kim Inchul, Cho Jaemin, Cho Hyeongjun, and Kim Min H.
Go with the Flow: Fluid Simulation and Rendering
Adaptive Optical Layers: Efficient Tall Cell Grids for Liquid Simulation
Fumiya Narita and Takashi Kanai
A Semi-Analytical Energy Model for Particle-Based Fluid Simulation Involving Complex Moving Boundaries
Junyuan Liu, Shusen Liu, Yuzhong Guo, Ruikai Liang, Yin Li, and Xiaowei He
Dripping Thin Films for Real-time Digital Painting
Herson Zoé, Paris Axel, and Michel Élie
Solving Deformation: Numerical Methods for Elastic Simulation
STAGED: Stress-Tensor Assisted Global-local-global solver for interactive Elastic shape Design
Ruan Liangwang, Wang Bin, Liu Tiantian, and Chen Baoquan
Interpolated Adaptive Linear Reduced Order Modeling for Deformation Dynamics
Tao Yutian, Chiaramonte Maurizio, and Fernandez Pablo
Progressively Projected Newton's Method
Fernández-Fernández José Antonio, Löschner Fabian, and Bender Jan
Affinification: A Fine Approximation of Deformations
Mercier-Aubin Alexandre, Schneider Teseo, Kry Paul, and Andrews Sheldon
Immersive and Interactive: Rendering Across Displays and Devices
Robo-Saber: Generating and Simulating Virtual Reality Players
Kim Nam Hee, Liu May, Lehtinen Jaakko, Hämäläinen Perttu, O'Brien James, and Peng Jason
Real-Time Neural Materials on Mobile VR
Zilin Xu, Yang Zhou, Yehonathan Litman, Matt Jen-Yuan Chiang, Lingqi Yan, and Anton Michels
ML-PEA: Machine Learning-Based Perceptual Algorithms for Display Power Optimization
Chen Kenneth, Matsuda Nathan, Wan Thomas, Ninan Ajit, Chapiro Alexandre, and Sun Qi
ProjectiveShading: Inserting 3D Objects into Indoor Images with Complex Shadows
Luo Jundan, Wu Xiaolong, Zhao Nanxuan, Wang Lu, Li Wenbin, and Richardt Christian
PBR-Inspired Controllable Diffusion for Image Generation
Xue Bowen, Guarnera Giuseppe Claudio, Zhao Shuang, and Montazeri Zahra

BibTeX (45-Issue 2)
                
@article{
10.1111:cgf.70316,
journal = {Computer Graphics Forum}, title = {{
Lightmap Compression with Color-Coherent UV Clustering and Cascade Texture Optimization}},
author = {
Chen, Dehan
and
Huang, Hongyu
and
Luo, Yuzhe
and
Xu, Hao
and
Zhang, Yuqing
and
Yang, Sipeng
and
Gao, Xifeng
and
Cai, Heng
and
Li, Chao
and
Jin, Xiaogang
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70316}
}
                
@article{
10.1111:cgf.70317,
journal = {Computer Graphics Forum}, title = {{
RotGS: Rotation-Guided 3D Gaussian Splatting for Turntable Sequences without Structure-from-Motion}},
author = {
Kim, Kyumin
and
Lee, Dohae
and
Baek, Hanul
and
Lee, In-Kwon
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70317}
}
                
@article{
10.1111:cgf.70320,
journal = {Computer Graphics Forum}, title = {{
ProjectiveShading: Inserting 3D Objects into Indoor Images with Complex Shadows}},
author = {
Luo, Jundan
and
Wu, Xiaolong
and
Zhao, Nanxuan
and
Wang, Lu
and
Li, Wenbin
and
Richardt, Christian
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70320}
}
                
@article{
10.1111:cgf.70319,
journal = {Computer Graphics Forum}, title = {{
Story2Board: A Training-Free Approach for Expressive Visual Storytelling}},
author = {
Dinkevich, David
and
Levy, Matan
and
Avrahami, Omri
and
Samuel, Dvir
and
Lischinski, Dani
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70319}
}
                
@article{
10.1111:cgf.70321,
journal = {Computer Graphics Forum}, title = {{
Statistical Denoising of Transient Rendering}},
author = {
Pueyo-Ciutad, Oscar
and
Lopez, Alvaro
and
Gutierrez, Diego
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70321}
}
                
@article{
10.1111:cgf.70325,
journal = {Computer Graphics Forum}, title = {{
GeoFusionLRM: Geometry-Aware Self-Correction for Consistent 3D Reconstruction}},
author = {
Yildirim, Ahmet Burak
and
Saygin, Tuna
and
Ceylan, Duygu
and
Dundar, Aysegul
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70325}
}
                
@article{
10.1111:cgf.70326,
journal = {Computer Graphics Forum}, title = {{
Basis Networks: Learning basis functions for free-form triangulations}},
author = {
Djuren, Tobias
and
Alexa, Marc
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70326}
}
                
@article{
10.1111:cgf.70327,
journal = {Computer Graphics Forum}, title = {{
EUROGRAPHICS 2026: CGF 45-2 Frontmatter}},
author = {}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70327}
}
                
@article{
10.1111:cgf.70334,
journal = {Computer Graphics Forum}, title = {{
ContactVision: Learning Foot Contact from Video for Physically Plausible Gait Animation}},
author = {
Kim, Daeyong
and
Yi, Gyuseok
and
Yu, Ri
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70334}
}
                
@article{
10.1111:cgf.70336,
journal = {Computer Graphics Forum}, title = {{
Physics-Based Motion Tracking of Contact-Rich Interacting Characters}},
author = {
Zhang, Xiaotang
and
Chang, Ziyi
and
Men, Qianhui
and
Shum, Hubert P. H.
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70336}
}
                
@article{
10.1111:cgf.70335,
journal = {Computer Graphics Forum}, title = {{
Vector sketch animation generation with differentiable motion trajectories}},
author = {
Zhu, Xinding
and
Yang, Xinye
and
Zheng, Shuyang
and
Zhang, Zhexin
and
Gao, Fei
and
Huang, Jing
and
Chen, Jiazhou
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70335}
}
                
@article{
10.1111:cgf.70332,
journal = {Computer Graphics Forum}, title = {{
TABI: Tight and Balanced Interactive Atlas Packing}},
author = {
Gu, Floria
and
Vining, Nicholas
and
Sheffer, Alla
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70332}
}
                
@article{
10.1111:cgf.70347,
journal = {Computer Graphics Forum}, title = {{
GS-2M: Material-aware Gaussian Splatting for High-fidelity Mesh Reconstruction}},
author = {
Nguyen, Dinh Minh
and
Avenhaus, Malte
and
Lindemeier, Thomas
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70347}
}
                
@article{
10.1111:cgf.70349,
journal = {Computer Graphics Forum}, title = {{
Volume Quantization with Flexible Singularities for Hexahedral Meshing}},
author = {
Brückler, Hendrik
and
Campen, Marcel
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70349}
}
                
@article{
10.1111:cgf.70354,
journal = {Computer Graphics Forum}, title = {{
Mesh Processing Non-Meshes via Neural Displacement Fields}},
author = {
Noma, Yuta
and
Wang, Zhecheng
and
Liu, Chenxi
and
Singh, Karan
and
Jacobson, Alec
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70354}
}
                
@article{
10.1111:cgf.70330,
journal = {Computer Graphics Forum}, title = {{
DexterCap: Affordable and Automated Capture of Complex Hand-Object Interactions}},
author = {
Liang, Yutong
and
Xu, Shiyi
and
Zhang, Yulong
and
Zhan, Bowen
and
Zhang, He
and
Liu, Libin
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70330}
}
                
@article{
10.1111:cgf.70365,
journal = {Computer Graphics Forum}, title = {{
Dance Like a Chicken: Low-Rank Stylization for Human Motion Diffusion}},
author = {
Sawdayee, Haim
and
Guo, Chuan
and
Tevet, Guy
and
Zhou, Bing
and
Wang, Jian
and
Bermano, Amit Haim
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70365}
}
                
@article{
10.1111:cgf.70370,
journal = {Computer Graphics Forum}, title = {{
Floorplan Generation by Alternating Geometry and Semantics Optimization}},
author = {
Wu, Wenming
and
Hu, Sizhe
and
Liu, Ligang
and
Zheng, Liping
and
Fu, Xiao-Ming
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70370}
}
                
@article{
10.1111:cgf.70375,
journal = {Computer Graphics Forum}, title = {{
Layer3D: A 3D Layered Representation for Multiview Vector Graphics}},
author = {
Guan, Zhongyue
and
Hu, Yixin
and
Wang, Zeyu
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70375}
}
                
@article{
10.1111:cgf.70377,
journal = {Computer Graphics Forum}, title = {{
VQ-Style: Disentangling Style and Content in Motion with Residual Quantized Representations}},
author = {
Zargarbashi, Fatemeh
and
Agrawal, Dhruv
and
Buhmann, Jakob
and
Guay, Martin
and
Coros, Stelian
and
Sumner, Robert W.
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70377}
}
                
@article{
10.1111:cgf.70378,
journal = {Computer Graphics Forum}, title = {{
UniCross3D: Unified Cross-View and Cross-Domain Diffusion for Consistent Single-Image 3D Generation}},
author = {
Jun, U-Chae
and
Ko, Jaeeun
and
Kang, Jiwoo
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70378}
}
                
@article{
10.1111:cgf.70379,
journal = {Computer Graphics Forum}, title = {{
DiskScissors: Cutting Arbitrary-Topology Solids for Bijective Mapping}},
author = {
Hinderink, Steffen
and
Campen, Marcel
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70379}
}
                
@article{
10.1111:cgf.70388,
journal = {Computer Graphics Forum}, title = {{
Skeletal-Driven Animation of Anatomical Humans via Neural Deformation Gradients}},
author = {
Nolte, Gerrit
and
Kemper, Fabian
and
Schwanecke, Ulrich
and
Botsch, Mario
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70388}
}
                
@article{
10.1111:cgf.70404,
journal = {Computer Graphics Forum}, title = {{
Hierarchical Optimization of the As-Rigid-As-Possible Energy}},
author = {
Meyer, Hendrik
and
Bickel, Bernd
and
Alexa, Marc
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70404}
}
                
@article{
10.1111:cgf.70411,
journal = {Computer Graphics Forum}, title = {{
Convex Primitive Decomposition for Collision Detection}},
author = {
Knodt, Julian
and
Gao, Xifeng
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70411}
}
                
@article{
10.1111:cgf.70412,
journal = {Computer Graphics Forum}, title = {{
Conversational Gesture Model (CGM): Extending Speaker-Centric Audio-Driven Motion Generation to Full Conversation Gestures}},
author = {
Koren, Tomer
and
Rosenthal, Adi
and
Friedman, Doron
and
Shamir, Ariel
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70412}
}
                
@article{
10.1111:cgf.70414,
journal = {Computer Graphics Forum}, title = {{
Mixed Super-Circles}},
author = {
Hohnadel, Emile
and
Métivet, Thibaut
and
Bertails-Descoubes, Florence
}, year = {
2026},
publisher = {
The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {
10.1111/cgf.70414}
}

Browse

Recent Submissions

Now showing 1 - 60 of 82
  • Item
    Lightmap Compression with Color-Coherent UV Clustering and Cascade Texture Optimization
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Chen, Dehan; Huang, Hongyu; Luo, Yuzhe; Xu, Hao; Zhang, Yuqing; Yang, Sipeng; Gao, Xifeng; Cai, Heng; Li, Chao; Jin, Xiaogang; Masia, Belen; Thies, Justus
    To address the storage overhead of lightmaps and the limitations of existing compression techniques, we propose a novel UV-space compression framework based on per-triangle processing. By mapping triangles to a standardized domain, we cluster and repack color-coherent regions into a compact atlas, generating a cascade texture refined via differentiable rendering. Experimental results show an average storage reduction of 83% with approximately 10 dB higher PSNR than existing methods. Our approach is the first dedicated lightmap compression framework compatible with standard block-based formats, offering an effective solution for memory-efficient 3D asset delivery.
  • Item
    RotGS: Rotation-Guided 3D Gaussian Splatting for Turntable Sequences without Structure-from-Motion
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Kim, Kyumin; Lee, Dohae; Baek, Hanul; Lee, In-Kwon; Masia, Belen; Thies, Justus
    The field of 3D reconstruction from multi-view images has advanced rapidly thanks to 3D Gaussian Splatting (3DGS), which enables efficient and photorealistic scene representation. However, optimizing 3DGS requires high-quality images from various viewpoints with accurate camera poses. The repeated collection of such data demands significant human effort, which poses a major constraint in practical applications. To address this issue, automated capturing systems that uses a turntable and fixed camera are widely employed. In a turntable setup, the background remains stationary while the object rotates. Therefore, preprocessing to remove the backgrond is essential, but the preprocessing reduces the number of reliable feature matches, which destabilizes Structure-from-Motion (SfM). This results in inaccurate camera poses, which degrades the quality of 3DGS reconstruction. We propose a novel method to optimize 3DGS in a turntable setup without SfM by leveraging the prior knowledge that objects rotate around a central axis. Unlike previous SfM-free methods that estimate camera poses for each frame, our approach reduces the complexity of optimization by representing rotations with a single global rotation axis. The estimated rotation is directly applied to the 3D Gaussians, producing motion defined as rotation flow. This rotation flow is then aligned with optical flow to provide strong geometric supervision. Through uncertainty-to-detail flow scheduling, our approach remains stable during the initial training stage when the geometry of the Gaussian set is still inaccurate. On the NeRF-Synthetic dataset and on real-world datasets captured with a turntable, our method outperforms existing SfM-free approaches in both reconstruction quality and training speed, and even demonstrates performance comparable to 3DGS optimized with precise camera poses.
  • Item
    ProjectiveShading: Inserting 3D Objects into Indoor Images with Complex Shadows
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Luo, Jundan; Wu, Xiaolong; Zhao, Nanxuan; Wang, Lu; Li, Wenbin; Richardt, Christian; Masia, Belen; Thies, Justus
    Realistically inserting virtual 3D objects into real-world images requires perceptually coherent shadowing of the object and background scene. Achieving this in single-view indoor scenes with sunlight is challenging due to complex, partially visible occluders and indirect lighting. Environment maps alone cannot produce realistic shadows on virtual objects, and any representation (scene parameters) used for rendering must be practically estimable. We introduce ProjectiveShading, the first automatic method for inverse- and re-rendering that handles bi-directional shadow interactions for realistic object composition. Our key innovation is the sunlight map, a 2D image encoding direct sunlight and arbitrary occlusions. It is generated from single-view estimations using off-the-shelf models and is compatible with standard rendering engines. We also propose algorithms to estimate sunlight direction and to blend virtual and real shadows while preserving background textures. Experiments on synthetic and in-the-wild images show our method outperforms previous approaches.
  • Item
    Story2Board: A Training-Free Approach for Expressive Visual Storytelling
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Dinkevich, David; Levy, Matan; Avrahami, Omri; Samuel, Dvir; Lischinski, Dani; Masia, Belen; Thies, Justus
    We present Story2Board, a training-free framework for expressive storyboard generation from natural language. Existing methods narrowly focus on subject identity, overlooking key aspects of visual storytelling such as spatial composition, background evolution, and narrative pacing. To address this, we introduce a lightweight consistency framework composed of two components: Latent Panel Anchoring, which preserves a shared character reference across panels, and Reciprocal Attention Value Mixing, which softly blends visual features between token pairs with strong reciprocal attention. Together, these mechanisms enhance coherence without architectural changes or fine-tuning, enabling state-of-the-art diffusion models to generate visually diverse yet consistent storyboards. To structure generation, we use an off-the-shelf language model to convert free-form stories into grounded panel-level prompts. To evaluate, we propose the Rich Storyboard Benchmark, a suite of open-domain narratives designed to assess layout diversity and background-grounded storytelling, in addition to consistency. We also introduce a new Scene Diversity metric that quantifies spatial and pose variation across storyboards. Our qualitative and quantitative results, as well as a user study, show that Story2Board produces more dynamic, coherent, and narratively engaging storyboards than existing baselines. Project page: https://daviddinkevich.github.io/Story2Board/
  • Item
    Statistical Denoising of Transient Rendering
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Pueyo-Ciutad, Oscar; Lopez, Alvaro; Gutierrez, Diego; Masia, Belen; Thies, Justus
    Transient rendering simulates light in motion, measuring the time of flight from the light source to the camera. However, the stochastic nature of Monte Carlo is aggravated in transient rendering, since samples are now spread along the temporal domain. In our work, we propose to denoise transient Monte Carlo renders by exploiting the spatio-temporal correlation of transient light transport, extending a recent statistical denoising formulation. By relying on statistics, we achieve a near-optimal tradeoff between reduced variance and introduced bias. We efficiently collect per-time-bin statistics in the temporal domain while avoiding impractical memory requirements, and use these collected statistics to analyze the spatio-temporal correlation and discriminate which time bins should be combined. Our statistics-based transient denoiser does not hallucinate, guarantees convergence of the result, is efficient, does not require any training and naturally handles participating media. We believe that the generality of our method might pave the way for denoising time-resolved Monte Carlo simulations in other domains, such as non-line-of-sight imaging, acoustic rendering, or absorption microscopy.
  • Item
    GeoFusionLRM: Geometry-Aware Self-Correction for Consistent 3D Reconstruction
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Yildirim, Ahmet Burak; Saygin, Tuna; Ceylan, Duygu; Dundar, Aysegul; Masia, Belen; Thies, Justus
    Single-image 3D reconstruction with large reconstruction models (LRMs) has advanced rapidly, yet reconstructions often exhibit geometric inconsistencies and misaligned details that limit fidelity. We introduce GeoFusionLRM, a geometry-aware selfcorrection framework that leverages the model's own normal and depth predictions to refine structural accuracy. Unlike prior approaches that rely solely on features extracted from the input image, GeoFusionLRM feeds back geometric cues through a dedicated transformer and fusion module, enabling the model to correct errors and enforce consistency with the conditioning image. This design improves the alignment between the reconstructed mesh and the input views without additional supervision or external signals. Extensive experiments demonstrate that GeoFusionLRM achieves sharper geometry, more consistent normals, and higher fidelity than state-of-the-art LRM baselines.
  • Item
    Basis Networks: Learning basis functions for free-form triangulations
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Djuren, Tobias; Alexa, Marc; Masia, Belen; Thies, Justus
    We present a framework for learning compactly supported basis functions that define tangent continuous surfaces based on coarse irregular triangle meshes. The basis functions are represented as MLPs. Smoothness of the basis functions is achieved by using the values of Loop basis functions as the parameterization of the surface. Post-multiplying the value of the MLP with the Loop basis yields smooth compact support. We show that this approach works similar or better than Neural Subdivision in terms of recreating given geometry, while the runtime scales better with surface resolution and can be evaluated at arbitrary resolution.
  • Item
    EUROGRAPHICS 2026: CGF 45-2 Frontmatter
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Masia, Belen; Thies, Justus
  • Item
    ContactVision: Learning Foot Contact from Video for Physically Plausible Gait Animation
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Kim, Daeyong; Yi, Gyuseok; Yu, Ri; Masia, Belen; Thies, Justus
    Foot-ground contact information plays a crucial role in character animation and gait analysis...
  • Item
    Physics-Based Motion Tracking of Contact-Rich Interacting Characters
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Zhang, Xiaotang; Chang, Ziyi; Men, Qianhui; Shum, Hubert P. H.; Masia, Belen; Thies, Justus
    Motion tracking has been an important technique for imitating human-like movement from large-scale datasets in physics-based motion synthesis. However, existing approaches focus on tracking either single character or a particular type of interaction, limiting their ability to handle contact-rich interactions. Extending single-character tracking approaches suffers from instability due to forces transferred through contacts. Contact-rich interactions require levels of control that place greater demands on model capacity. To this end, we propose a robust tracking method based on progressive neural networks (PNN) where multiple experts specialize in learning skills of various difficulties. Our method automatically assigns training samples to experts without manual scheduling. Both qualitative and quantitative results show that our method delivers more stable motion tracking in densely interactive movements while enabling more efficient model training.
  • Item
    Vector sketch animation generation with differentiable motion trajectories
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Zhu, Xinding; Yang, Xinye; Zheng, Shuyang; Zhang, Zhexin; Gao, Fei; Huang, Jing; Chen, Jiazhou; Masia, Belen; Thies, Justus
    Sketching is a direct and inexpensive means of visual expression...
  • Item
    A Real-Time Multi-Scale Neural Representation for Complex Surface Reflectance
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Timonen, Heikki; Kemppinen, Pauli; Lehtinen, Jaakko; Masia, Belen; Thies, Justus
    Recent machine learning methods have significantly advanced the state of the art in the classic problem of representing surface appearance over angle, space, and scale. The models tend, however, to be relatively heavy compared to traditional fixedfunction representations, making real-time application challenging. We present a neural shading architecture that allows the use of smaller and faster-to-evaluate neural networks than current state of the art, while faithfully representing complex spatial and angular variation. We target the angular complexity that arises both from prefiltering normal-mapped SVBRDFs, as well as complex, measured homogeneous BRDFs. A key architectural innovation is the introduction of a multiplicative interaction ("gating") between learnable parameters that significantly increases our model's expressive power. Our straightforward, unoptimized shader implementation renders over 1000 full HD frames per second on a consumer GPU using our default parameters.
  • Item
    TABI: Tight and Balanced Interactive Atlas Packing
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Gu, Floria; Vining, Nicholas; Sheffer, Alla; Masia, Belen; Thies, Justus
    Atlas packing is a key step in many computer graphics applications. Packing algorithms seek to arrange a set of charts within a fixed-size atlas with as little downscaling as possible. Many packing applications such as content creation tools, dynamic atlas generation for video games, and texture space shading require on-the-fly interactive atlas packing. Unfortunately, while many methods have been developed for generating tight high-quality packings, they are designed for offline settings and have running times two or more orders of magnitude greater than what is required for interactive performance. While real-time GPU packing methods exist, they significantly downscale packed charts compared to offline methods. We introduce a GPU packing method that targets interactive speeds, provides packing quality approaching that of offline methods, and supports flexible user control over the tradeoff between performance and quality. We observe that current real-time packing methods leave large gaps between charts and often produce asymmetric, or poorly balanced, packings. These artifacts dramatically degrade packing quality. Our Tight And Balanced method eliminates these artifacts while retaining Interactive performance. TABI generates tight packings by compacting empty space between irregularly shaped charts both horizontally and vertically, using two approximations of chart shape that support efficient parallel processing. We balance packing outputs by automatically adjusting atlas row widths and orientations to accommodate varying chart heights. We show that our method significantly reduces chart downscaling compared to existing interactive methods while remaining orders of magnitude faster than offline alternatives.
  • Item
    Variable-Rate Texture Compression: Real-Time Rendering with JPEG
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Kristmann, Elias; Wimmer, Michael; Schütz, Markus; Masia, Belen; Thies, Justus
    Although variable-rate compressed image formats such as JPEG are widely used to efficiently encode images, they have not found their way into real-time rendering due to special requirements such as random access to individual texels. In this paper, we investigate the feasibility of variable-rate texture compression on modern GPUs using the JPEG format, and how it compares to the GPU-friendly fixed-rate compression approaches BC1 and ASTC. Using a deferred rendering pipeline, we are able to identify the subset of blocks that are needed for a given frame, decode these, and colorize the framebuffer's pixels. Despite the additional ∼0.17 bit per pixel that we require for our approach, JPEG maintains significantly better quality and compression rates compared to BC1, and depending on the type of image, outperforms or competes with ASTC. The JPEG rendering pipeline increases rendering duration by less than 0.3 ms on an NVIDIA RTX 4090, demonstrating that sophisticated variable-rate compression schemes are feasible on modern GPUs, even in VR. Source code and data sets are available at: https://github.com/elias1518693/jpeg_textures
  • Item
    Splat-based Metal Artifact Reduction in Cone-Beam CT via Polychromatic Modeling
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Choi, Kiseok; Kim, Inchul; Cho, Jaemin; Cho, Hyeongjun; Kim, Min H.; Masia, Belen; Thies, Justus
    Cone-beam computed tomography (CBCT) enables volumetric reconstruction from X-ray projections, but suffers from severe artifacts-especially beam hardening-when imaging materials with high attenuation such as metals. These artifacts arise from the polychromatic nature of X-rays and are not properly addressed by conventional monochromatic reconstruction algorithms. While recent neural representation-based methods offer improved reconstruction quality, they are computationally expensive and often impractical for deployment. We propose a novel physics-inspired, self-calibrating metal artifact reduction method that efficiently reconstructs 3D CBCT volumes while correcting beam hardening artifacts. Our method integrates a polychromatic X-ray projection model, material-dependent attenuation profiles, and system response modeling into a Gaussian Splatting framework. Unlike prior work, we eliminate the need for manual metal masks or strong prior assumptions, and we optimize both reconstruction parameters and X-ray spectral characteristics jointly during training. We further introduce a high-fidelity synthetic CBCT dataset generation pipeline validated on Monte-Carlo x-ray simulation toolbox and release new datasets with severe metal-induced artifacts to support the community. This is the first splat-based method for reducing beam hardening in CBCT. Extensive experiments on both synthetic and real-world datasets demonstrate that our method outperforms state-of-the-art approaches in artifact suppression and reconstruction accuracy.
  • Item
    2D Piecewise Linear Scalar Fields with Invertible Integral Lines
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Erxleben, Timm Leon; Motejat, Michael; Rössl, Christian; Theisel, Holger; Masia, Belen; Thies, Justus
    Integral lines of the gradient flow are standard features in continuously differentiable scalar fields that enjoy some useful properties: They cover the domain densely, do not split, merge, or intersect, and are therefore invertible. For widely used discretizations of scalar fields, the corresponding polygonal approximations of integral lines do not enjoy these properties anymore. We analyze conditions for integral lines in 2D piecewise linear (PL) scalar fields to be invertible by identifying and classifying critical edges in the underlying triangulation. We show that under mild conditions, every 2D PL scalar field can be transformed into an arbitrarily close PL field with invertible integral lines. We present an algorithm that computes this transformation and apply it to a number of test data sets.
  • Item
    TextFlux: An OCR-Free DiT Model for High-Fidelity Multilingual Scene Text Synthesis
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Xie, Yu; Zhang, Jielei; Chen, Pengyu; Wang, Weihang; Gao, Longwen; Li, Peiyi; Qiao, Qian; Lian, Zhouhui; Masia, Belen; Thies, Justus
    Diffusion-based scene text synthesis has progressed rapidly...
  • Item
    SEE4D: Pose-Free 4D Generation via Auto-Regressive Video Inpainting
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Lu, Dongyue; Liang, Ao; Huang, Tianxin; Fu, Xiao; Zhao, Yuyang; Ma, Baorui; Pan, Liang; Yin, Wei; Kong, Lingdong; Ooi, Wei Tsang; Liu, Ziwei; Masia, Belen; Thies, Justus
    Immersive applications call for synthesizing spatiotemporal 4D content from casual videos without costly 3D supervision...
  • Item
    GS-2M: Material-aware Gaussian Splatting for High-fidelity Mesh Reconstruction
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Nguyen, Dinh Minh; Avenhaus, Malte; Lindemeier, Thomas; Masia, Belen; Thies, Justus
    We propose a material-aware optimization framework for high-fidelity mesh reconstruction...
  • Item
    Volume Quantization with Flexible Singularities for Hexahedral Meshing
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Brückler, Hendrik; Campen, Marcel; Masia, Belen; Thies, Justus
    We present a novel algorithm for quantization and subsequent hexahedral mesh generation from seamless volumetric maps. Quantization is the process of choosing integers that represent the numbers of hexahedral elements to be placed in each region of the volume, and transforming the seamless map into an integer-grid map matching that choice, inducing a hexahedral mesh. Previous work computes such quantizations under the restriction of a fixed predetermined singularity graph. Our novel approach allows for implicit modification and, in particular, simplification of the map's singularity structure wherever that benefits the chosen objective, such as matching target hexahedron sizes as closely as possible. It comes with two novel ingredients: A feature-focused distortion measure guiding the quantization, and constraints ensuring map injectivity and structure preservation of geometric and topological features, both without relying on a fixed singularity structure. We demonstrate the benefit of the added flexibility offered by this approach: it allows for the generation of hexahedral meshes that more accurately match a desired resolution globally, as well as of meshes exhibiting a simpler block structure.
  • Item
    Designing inflatable shells using unstructured meshes
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) He, Siyuan; Lebée, Arthur; Skouras, Mélina; Masia, Belen; Thies, Justus
    We propose a method for designing inflatable shells made of superimposed quasi-inextensible membranes sealed according to specific welding patterns. The shapes of the patterns are defined on unit triangular cells and allow us to locally control the contraction of the cells from rest to inflated configuration. By paving the patterns on the surface of the structure and properly grading their parameters, we are able to generate inflatables of prescribed deployed shapes through metric frustration. Our triangular cells exhibit isotropic contraction. Our inverse design algorithm can thus leverage conformal parametrization to compute local contraction ratios that are converted to pattern geometries. Key to our approach is to define the arrangement of the patterns using an unstructured triangulation of the target surface. Compared to more traditional arrangements laid out on regular grids, our approach allows us to easily cut or segment the structure without visible seams and to decrease the contraction range necessary to reproduce a given surface, thus enlarging our design space. Additionally, unstructured arrangements lead to inflatable structures whose cells conform to the boundary of the target surface and therefore better covers it. We demonstrate the capabilities of our approach in simulation and by fabricating various prototypes, made of one or multiple components.
  • Item
    CADrawer: Autoregressive CAD Generation from 3D Sketches
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Li, Yuanbo; Manfredi, Gilda; Kriel, Henro; Hao, Chengye; Xu, Xianghao; Bousseau, Adrien; Ritchie, Daniel; Masia, Belen; Thies, Justus
    In professional design workflows, designers often begin by creating sketch drawings before converting them into CAD programs. However, prior work on automatically interpreting these sketches has been limited to simplified inputs and fails to account for construction lines that are ubiquitous in real-world drawings. We present CADrawer, a system that translates 3D sketches into CAD programs using an autoregressive approach, leveraging construction lines as a rich source of information for recovering intermediate CAD operations. At each step, CADrawer predicts the next modeling operation and its parameters based on a graph-based representation of the sketch, which explicitly encodes spatial and temporal relationships between strokes. To improve generation quality, the system maintains multiple candidate programs in parallel, and a learned value function evaluates these partial programs to guide the search toward the most promising candidates. CADrawer is designed as a complement to 3D sketching interfaces, building on existing methods that create 3D sketches. We evaluate our method across several datasets, including those containing dense construction lines and cases without ground-truth B-rep shapes.
  • Item
    Mesh Processing Non-Meshes via Neural Displacement Fields
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Noma, Yuta; Wang, Zhecheng; Liu, Chenxi; Singh, Karan; Jacobson, Alec; Masia, Belen; Thies, Justus
    Mesh processing pipelines are mature, but adapting them to newer non-mesh surface representations—which enable fast rendering with compact file size—requires costly meshing or transmitting bulky meshes, negating their core benefits for streaming applications. We present a compact neural field that enables common geometry processing tasks across diverse surface representations. Given an input surface, our method learns a neural map from its coarse mesh approximation to the surface. The full representation totals only a few hundred kilobytes, making it ideal for lightweight transmission. Our method enables fast extraction of manifold and Delaunay meshes for intrinsic shape analysis, and compresses scalar fields for efficient delivery of costly precomputed results. Experiments and applications show that our fast, compact, and accurate approach opens up new possibilities for interactive geometry processing.
  • Item
    A Texture-Free Multi-Scale Model for Surface-Based Rendering of Knitted Fabrics
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Khattar, Apoorv; Aubry, Jean-Marie; Yan, Ling-Qi; Montazeri, Zahra; Masia, Belen; Thies, Justus
    Knitted fabrics present unique challenges for realistic rendering due to their complicated structure and scale-dependent appearance. Existing methods typically rely on explicit yarn geometry, which is computationally complex, or texture-based representations that require heavy storage and precomputed maps. In this paper, we introduce the first texture-free, surface-based appearance model for knitted fabrics, in which stitches are represented parametrically as thick curves and mapped directly onto fabric meshes. This avoids explicit yarn or fiber geometry, yet preserves the characteristic 3D look of yarn-based models. Unlike prior surface-based approaches, our method produces realistic volumetric effects such as depth, parallax, and silhouette preservation. To achieve this, we propose a curvature-aware parallax mapping technique that ensures coherent appearance at grazing angles. Furthermore, we extend the appearance model to a multi-scale formulation that aggregates geometry and visibility over texture footprints and adjusts roughness parameters for stable far-field rendering. Our model combines the efficiency and simplicity of surface-based methods with the volumetric realism of fiber-based models, reproducing characteristic knit effects such as 3D stitch structure in a multi-scale manner without the complexity or storage cost of texture-based approaches.
  • Item
    TLC-Plan: A Two-Level Codebook Based Network for End-to-End Vector Floorplan Generation
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Xiong, Biao; Peng, Zhen; Wang, Ping; Liu, Qiegen; Zhong, Xian; Masia, Belen; Thies, Justus
    Automated floorplan generation aims to improve design quality, architectural efficiency, and sustainability by jointly modeling global spatial organization and precise geometric detail. However, existing approaches operate in raster space and rely on post hoc vectorization, which introduces structural inconsistencies and hinders end-to-end learning. Motivated by compositional spatial reasoning, we propose TLC-Plan, a hierarchical generative model that directly synthesizes vector floorplans from input boundaries, aligning with human architectural workflows based on modular and reusable patterns. TLC-Plan employs a two-level VQ-VAE to encode global layouts as semantically labeled room bounding boxes and to refine local geometries using polygon-level codes. This hierarchy is unified in a CodeTree representation, while an autoregressive transformer samples codes conditioned on the boundary to generate diverse and topologically valid designs without requiring explicit room topology or dimensional priors. Extensive experiments show state-of-the-art performance on the RPLAN dataset and strong results on the LIFULL dataset. The proposed framework advances constraint-aware and scalable vector floorplan generation for real-world architectural applications.
  • Item
    Field-Aligned Surface-Filling Curve via Implicit Stitching
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Cocco, Giovanni; Chermain, Xavier; Masia, Belen; Thies, Justus
    We present a robust and scalable method for generating field-aligned surface-filling curves on general manifolds. Building upon prior work on stripe pattern generation and field-aligned surface-filling curves, our approach introduces a novel stitching strategy that operates directly in the implicit domain. Unlike previous methods that extract and stitch isolines post hoc, we perform stitching by manipulating the scalar field itself, enabling an efficient and robust solution that generalizes beyond planar surfaces. We demonstrate more than an order of magnitude speed-up and improved alignment with input direction fields when compared to state-of-the-art geometric flow methods. Robustness is validated on the Thingi10K dataset. Moreover, the method integrates with Blender 4.5 for interactive curve generation on small models, and scales to massive meshes, producing surface-filling curves with over ten million vertices in under twenty-five minutes.
  • Item
    MultiCOIN: Multi-Modal COntrollable INbetweening
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Tanveer, Maham; Zhou, Yang; Niklaus, Simon; Mahdavi Amiri, Ali; Zhang, Hao (Richard); Singh, Krishna Kumar; Zhao, Nanxuan; Masia, Belen; Thies, Justus
    Video inbetweening creates smooth transitions between two frames making it an indispensable tool for video editing and longform video synthesis. Existing methods struggle with large or complex motion and offer limited control over intermediate frames, often misaligning with user intent. We introduce MultiCOIN, a video inbetweening framework supporting multi-modal controls, including depth transitions and layering, motion trajectories, text prompts, and target regions for movement localization. It balances flexibility, usability, and fine-grained precision. Built on a Diffusion Transformer (DiT), due to its proven capability to generate high-quality long video, our model maps all motion controls into a unified sparse point-based representation compatible with the denoising process. Further, to respect the variety of controls which operate at varying levels of granularity and influence, we separate content and motion into two branches, enabling dedicated generators for each. A stage-wise training strategy ensures stable learning of multi-modal controls. Extensive experiments show improved motion complexity, controllability, and narrative consistency. Project Page: MultiCOIN.
  • Item
    DexterCap: Affordable and Automated Capture of Complex Hand-Object Interactions
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Liang, Yutong; Xu, Shiyi; Zhang, Yulong; Zhan, Bowen; Zhang, He; Liu, Libin; Masia, Belen; Thies, Justus
    Capturing fine-grained hand-object interactions is challenging due to severe self-occlusion from closely spaced fingers and the subtlety of in-hand manipulation motions. Existing optical motion capture systems rely on expensive camera setups and extensive manual post-processing, while low-cost vision-based methods often suffer from reduced accuracy and reliability under occlusion. To address these challenges, we present DexterCap, a low-cost optical capture system for dexterous in-hand manipulation. DexterCap uses dense, character-coded marker patches to achieve robust tracking under severe self-occlusion, together with an automated reconstruction pipeline that requires minimal manual effort. With DexterCap, we introduce DexterHand, a dataset of fine-grained hand-object interactions covering diverse manipulation behaviors and objects, from simple primitives to complex articulated objects such as a Rubik's Cube. We release the dataset and code to support future research on dexterous hand-object interaction. Project website: https://pku-mocca.github.io/Dextercap-Page/
  • Item
    OUGS: Active View Selection via Object-aware Uncertainty Estimation in 3DGS
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Li, Haiyi; Chen, Qi; Kalkofen, Denis; Chen, Hsiang-Ting; Masia, Belen; Thies, Justus
    Recent advances in 3D Gaussian Splatting (3DGS) have achieved state-of-the-art results for novel view synthesis. However, efficiently capturing high-fidelity reconstructions of specific objects within complex scenes remains a significant challenge. A key limitation of existing active reconstruction methods is their reliance on scene-level uncertainty metrics, which are often biased by irrelevant background clutter and lead to inefficient view selection for object-centric tasks. We present OUGS, a novel framework that addresses this challenge with a more principled, physically-grounded uncertainty formulation for 3DGS. Our core innovation is to derive uncertainty directly from the explicit physical parameters of the 3D Gaussian primitives (e.g., position, scale, rotation). By propagating the covariance of these parameters through the rendering Jacobian, we establish a highly interpretable uncertainty model. This foundation allows us to seamlessly integrate semantic segmentation masks to produce a targeted, object-aware uncertainty score that effectively disentangles the object from its environment. This enables a more effective active view selection strategy that prioritizes views critical to improving object fidelity. Experimental evaluations on public datasets demonstrate that our approach significantly improves the efficiency of the 3DGS reconstruction process and achieves higher quality for targeted objects compared to existing state-of-the-art methods, while also serving as a robust uncertainty estimator for the global scene.
  • Item
    Graph-based Black and White Stylization
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Sattari Javid, Ali; Lord, Jimmy; Mould, David; Masia, Belen; Thies, Justus
    Stippling is an art style forming images from large numbers of small dots. Computer graphics practitioners have proposed numerous algorithms for stippling and stippling variants over the years. In this paper, we propose a black and white image stylization technique in which the image is formed from heterogeneous black and white vector marks, with stipples and larger regions represented by polygons. Our algorithm applies two passes of stippling: first, we apply stippling to an importance map of the input image; then, using the stipples from the first pass as nodes in a planar graph, we create a binary stippling by labeling each graph node as either black or white using error diffusion. A connected component analysis of the resulting labeling, followed by vectorization of the components, yields a high-quality black and white representation of the original image. Isolated nodes, sharing a color with none of their neighbors, are akin to stipples; larger groups are polygons. We provide quantitative and qualitative comparisons with previous work on stippling and black and white rendering.
  • Item
    Dance Like a Chicken: Low-Rank Stylization for Human Motion Diffusion
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Sawdayee, Haim; Guo, Chuan; Tevet, Guy; Zhou, Bing; Wang, Jian; Bermano, Amit Haim; Masia, Belen; Thies, Justus
    Text-to-motion generative models span a wide range of 3D human actions but struggle with nuanced stylistic attributes such as a 'Chicken' style. Due to the scarcity of style-specific data, existing approaches pull the generative prior towards a reference style, which often results in out-of-distribution, low-quality generations. In this work, we introduce LoRA-MDM, a lightweight framework for motion stylization that generalizes to complex actions while maintaining editability. Our key insight is that adapting the generative prior to include the style, while preserving its overall distribution, is more effective than modifying each individual motion during generation. Building on this idea, LoRA-MDM learns to adapt the prior to include the reference style using only a few samples. The style can then be used in the context of different textual prompts for generation. The low-rank adaptation shifts the motion manifold in a semantically meaningful way, enabling realistic style infusion even for actions not present in the reference samples. Moreover, preserving the distribution structure enables advanced operations such as style blending and motion editing. We compare LoRA-MDM to state-of-the-art stylized motion generation methods and demonstrate a favorable balance between text fidelity and style consistency.
  • Item
    TreeON: Reconstructing 3D Tree Point Clouds from Orthophotos and Heightmaps
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Grammatikaki, Angeliki; Eschner, Johannes; Hermosilla, Pedro; Argudo, Oscar; Waldner, Manuela; Masia, Belen; Thies, Justus
    We present TreeON, a novel neural-based framework for reconstructing detailed 3D tree point clouds from sparse top-down geodata, using only a single orthophoto and its corresponding Digital Surface Model (DSM). Our method introduces a new training supervision strategy that combines both geometric supervision and a differentiable shadow and silhouette loss to learn point cloud representations of trees without requiring species labels, procedural rules, detailed terrestrial reconstruction data, or ground laser scan data. To address the lack of ground truth data, we generate a synthetic dataset of point clouds from procedurally modeled trees and train our network on it. Quantitative and qualitative experiments demonstrate better reconstruction quality and coverage compared to existing methods, as well as strong generalization to real-world data, leading to visually appealing and structurally plausible tree point cloud representations that can be integrated into interactive digital 3D maps.
  • Item
    Strain-Field Based Segmentation for Fabric Formwork
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Sati, Abhinit; Bao, Tiffany; Tedi, Jeff; Chien, Edward; Whiting, Emily; Masia, Belen; Thies, Justus
    We present a physically-informed segmentation pipeline for producing fabric formwork for the casting and molding of arbitrary 3D objects.
  • Item
    Enhancing Robust Category-Agnostic Pose Estimation through Multi-Modal Feature Alignment
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Li, Boxuan; Liu, Juan; Masia, Belen; Thies, Justus
    Category-Agnostic Pose Estimation (CAPE) aims to detect keypoints for objects of any category using only a few labeled samples, making it a challenging yet crucial task for general-purpose visual understanding. Existing methods rely on either visual or textual inputs, but the lack of cross-modal interaction limits generalization. Without a unified input representation, solely using visual features hinders consistent prediction of same-type keypoints, while fixed textual representations fail to capture the diverse characteristics of same-type keypoints, leading to coarse and over-generalized outputs. To address these limitations, we propose two multi-modal frameworks that perform visual-textual integration at both the feature and decision levels. Our feature-level module leverages cross-modal attention to align and enhance keypoint representations, while the decision-level fusion adaptively combines modality-specific predictions through a modality-consistency loss. Experiments on the large-scale MP-100 dataset demonstrate that our method surpasses existing baselines in both accuracy and robustness. Under the challenging 1-shot setting, our model achieves a 0.58% improvement in PCK0.2 over the state-of-the-art CAPE method.
  • Item
    ML-PEA: Machine Learning-Based Perceptual Algorithms for Display Power Optimization
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Chen, Kenneth; Matsuda, Nathan; Wan, Thomas; Ninan, Ajit; Chapiro, Alexandre; Sun, Qi; Masia, Belen; Thies, Justus
    Image processing techniques can be used to modulate the pixel intensities of an image to reduce the power consumption of the display device. A simple example of this consists of uniformly dimming the entire image. Such algorithms should strive to minimize the impact on image quality while maximizing power savings. Techniques based on heuristics or human perception have been proposed, both for traditional flat panel displays and modern display modalities such as virtual and augmented reality (VR/AR). In this paper, we focus on developing and evaluating display power-saving techniques that use machine learning (ML) in VR displays. We developed a U-Net-based technique paired with perceptual and power optimization loss functions that generates spatially varying dimming maps. These dimming maps are used to modulate input images, per-pixel, to generate a power-efficient image. Our pipeline was validated via quantitative analysis using image quality metrics and through a subjective study. Our subjective validation provides results scaled in perceptual just-objectionable-difference (JOD) units. This data, when rescaled, allows for comparisons of our technique with recent studies on VR display power optimization. Our results show that participants prefer our technique over a uniform dimming baseline for high target power saving conditions.
  • Item
    Floorplan Generation by Alternating Geometry and Semantics Optimization
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Wu, Wenming; Hu, Sizhe; Liu, Ligang; Zheng, Liping; Fu, Xiao-Ming; Masia, Belen; Thies, Justus
    Creating floorplans lays the foundation for architectural design and scene modeling. We propose a novel framework for generating diverse high-quality floorplans under predefined constraints. Central to our method is an iterative refinement process for optimizing the bounding boxes of rooms and the floorplan semantics image, which defines a vector floorplan together. Vector floorplans can be generated through a learning-based refinement process. Our framework supports various constraints, such as floorplan boundaries, topological graphs, and bubble diagrams. Extensive experiments demonstrate that our method is superior to state-of-the-art techniques, particularly in generating a wider variety of solutions that cater to various architectural needs.
  • Item
    Interpolated Adaptive Linear Reduced Order Modeling for Deformation Dynamics
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Tao, Yutian; Chiaramonte, Maurizio; Fernandez, Pablo; Masia, Belen; Thies, Justus
    Linear reduced-order modeling (ROM) is widely used for efficient simulation of deformation dynamics, but its accuracy is often limited by the fixed linearization of the reduced mapping. We propose a new adaptive strategy for linear ROM that allows the reduced mapping to vary dynamically in response to the evolving deformation state, significantly improving accuracy over traditional linear approaches. To further handle large deformations, we introduce a historical displacement basis combined with Grassmann interpolation, enabling the system to recover robustly even in challenging scenarios. We evaluate our method through quantitative online-error analysis and qualitative comparisons with principal component analysis (PCA)-based linear ROM simulations, demonstrating substantial accuracy gains while preserving comparable computational costs.
  • Item
    LeafFit: Plant Assets Creation from 3D Gaussian Splatting
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Luo, Chang; Umetani, Nobuyuki; Masia, Belen; Thies, Justus
    We propose LeafFit, a pipeline that converts 3D Gaussian Splatting (3DGS) of individual plants into editable, instanced mesh assets. While 3DGS faithfully captures complex foliage, its high memory footprint and lack of mesh topology make it incompatible with traditional game production workflows. We address this by leveraging the repetition of leaf shapes; our method segments leaves from the unstructured 3DGS, with optional user interaction included as a fallback. A representative leaf group is selected and converted into a thin, sharp mesh to serve as a template; this template is then fitted to all other leaves via differentiable Moving Least Squares (MLS) deformation. At runtime, the deformation is evaluated efficiently on-the-fly using a vertex shader to minimize storage requirements. Experiments demonstrate that LeafFit achieves higher segmentation quality and deformation accuracy than recent baselines while significantly reducing data size and enabling parameter-level editing. Our source code is publicly available at https://github.com/netbeifeng/leaf_fit.
  • Item
    Layer3D: A 3D Layered Representation for Multiview Vector Graphics
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Guan, Zhongyue; Hu, Yixin; Wang, Zeyu; Masia, Belen; Thies, Justus
    We present Layer3D, a novel 3D neural representation that models objects as collections of decomposable neural implicit primitives. These primitives enable the generation of layered images with consistent correspondences across viewpoints, establishing a flexible framework for multiview vector graphics decomposition. Integrated into a text-to-3D pipeline via Score Distillation Sampling (SDS), Layer3D learns to generate primitives with diverse shape topologies while preserving structural coherence. To ensure front-to-back ordering required for 2D flat graphics, our method incorporates front-to-back rendering and shape regularization constraints. Experimental results demonstrate that Layer3D consistently produces meaningful, topology-diverse layers across multiple views, thereby facilitating intuitive and effective layer-based vector editing.
  • Item
    HeatMat: Simulation of City Material Impact on Urban Heat Island Effect
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Reinbigler, Marie; Rouffet, Romain; Naylor, Peter; Czerkawski, Mikolaj; Dionelis, Nikolaos; Brunet, Elisabeth; Fetita, Catalin; Martin, Rosalie; Masia, Belen; Thies, Justus
    The Urban Heat Island (UHI) effect, defined as a significant increase in temperature in urban environments compared to surrounding areas, is difficult to study in real cities using sensor data (satellites or in-situ stations) due to their coarse spatial and temporal resolution. Among the factors contributing to this effect are the properties of urban materials, which differ from those in rural areas. To analyze their individual impact and to test new material configurations, a high-resolution simulation at the city scale is required. Estimating the current materials used in a city, including those on building facades, is also challenging. We propose HeatMat, an approach to analyze at high resolution the individual impact of urban materials on the UHI effect in a real city, relying only on open data. We estimate building materials using street-view images and a pre-trained vision-language model (VLM) to supplement existing OpenStreetMap data, which describes the 2D geometry and features of buildings. We further encode this information into a set of 2D maps that represent the city's vertical structure and material characteristics. These maps serve as inputs for our 2.5D simulator, which models coupled heat transfers and enables random-access surface temperature estimation at multiple resolutions, reaching a 20x speedup compared to an equivalent simulation in 3D.
  • Item
    VQ-Style: Disentangling Style and Content in Motion with Residual Quantized Representations
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Zargarbashi, Fatemeh; Agrawal, Dhruv; Buhmann, Jakob; Guay, Martin; Coros, Stelian; Sumner, Robert W.; Masia, Belen; Thies, Justus
    Human motion data is inherently rich and complex, containing both semantic content and subtle stylistic features that are challenging to model. We propose a novel method for effective disentanglement of the style and content in human motion data to facilitate style transfer. Our approach is guided by the insight that content corresponds to coarse motion attributes while style captures the finer, expressive details. To model this hierarchy, we employ Residual Vector Quantized Variational Autoencoders (RVQ-VAEs) to learn a coarse-to-fine representation of motion. We further enhance the disentanglement by integrating codebook learning with contrastive learning and a novel information leakage loss to organize the content and the style across different codebooks. We harness this disentangled representation using our simple and effective inference-time technique Quantized Code Swapping, which enables motion style transfer without requiring any fine-tuning for unseen styles. Our framework demonstrates strong versatility across multiple inference applications, including style transfer, style removal, and motion blending.
  • Item
    UniCross3D: Unified Cross-View and Cross-Domain Diffusion for Consistent Single-Image 3D Generation
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Jun, U-Chae; Ko, Jaeeun; Kang, Jiwoo; Masia, Belen; Thies, Justus
    Reconstructing detailed geometry and realistic appearance from a single RGB image is essential yet fundamentally challenging due to inherent ambiguities such as occlusion, lighting variations, and texture-geometry entanglement. While recent diffusionbased generative models have significantly improved novel view synthesis, existing approaches suffer from two critical limitations: lack of cross-view geometric consistency and insufficient cross-domain semantic alignment. To address these issues, we introduce UNICROSS3D, a unified cross-view and cross-domain diffusion framework designed explicitly for consistent and physically coherent 3D generation. UNICROSS3D features two novel contributions: (1) a cross-view latent regularization that enforces cross-view geometric consistency across synthesized viewpoints by penalizing latent variance, and (2) a cross-domain mutual information objective grounded in the physics of image formation, explicitly aligning synthesized color and normal maps. Extensive experiments demonstrate that UNICROSS3D achieves significantly improved view consistency and semantic alignment over state-of-the-art methods and yields higher-fidelity reconstructions, particularly under challenging textures and ambiguous viewpoints.
  • Item
    DiskScissors: Cutting Arbitrary-Topology Solids for Bijective Mapping
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Hinderink, Steffen; Campen, Marcel; Masia, Belen; Thies, Justus
    An algorithm for cutting solid objects in a topology-controlled manner is presented. Concretely, given a loop on the object boundary, a disk-topology cut surface bounded by the loop is constructed in the interior. In contrast to various previous approaches, both disk topology and conformance to the prescribed loop are ensured by construction, while supporting not only contractible but also incontractible loops on the boundaries of manifold objects of higher genus and arbitrary non-trivial topology. We describe an implementation of this algorithm in the discrete setting, with triangle mesh cut surfaces embedded in tetrahedral mesh objects. Making use of this novel cutting algorithm, we describe a method for the reliable construction of bijective volumetric maps between solid objects, demonstrating the algorithm's utility. This mapping method overcomes restrictions of the state of the art to topological balls, extending coverage to objects of arbitrary genus, specifically so-called 1-handlebodies.
  • Item
    Affinification: A Fine Approximation of Deformations
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Mercier-Aubin, Alexandre; Schneider, Teseo; Kry, Paul; Andrews, Sheldon; Masia, Belen; Thies, Justus
    We introduce affinification, a novel method for accelerating physics-based animation of elastic solids. During a time-dependent simulation, our method automatically partitions the space into affine and elastic regions depending on the deformation. As such, we capture localized deformations while significantly reducing computational costs with larger regions of model reduction. We design a new clustering method based on deformation rates to capture affinely deforming regions, and explore multiple heuristics for seeding, pattern generation, and the impact of physical parameters on coarsened regions. We compare our method with the ground truth, showing performance increasing with resolution and recorded simulations up to 17× faster compared to elastic simulations, while retaining similar levels of visual fidelity.
  • Item
    Edge-preserving noise for diffusion models
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Vandersanden, Jente; Holl, Sascha; Huang, Xingchang; Singh, Gurprit; Masia, Belen; Thies, Justus
    Classical diffusion models typically rely on isotropic Gaussian noise, treating all regions uniformly and overlooking structural information important for high-quality generation. We introduce an edge-preserving diffusion process that generalizes isotropic models via a hybrid noise scheme with an edge-aware scheduler that smoothly transitions from edge-preserving to isotropic noise. This enables the model to capture fine structural details while generally maintaining global performance. We evaluate the impact of structure-aware noise in both diffusion and flow-matching frameworks, and show that existing isotropic models can be efficiently fine-tuned with edge-preserving noise, making our framework practical for adapting pre-trained systems. Beyond unconditional generation, our method particularly shows improvements in structure-guided tasks such as stroke-to-image synthesis, improving robustness and perceptual quality, as evidenced by consistent improvements across FID, KID, and CLIP-score.
  • Item
    Palette Aligned Image Diffusion
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Aharoni, Elad; Porat, Noy; Lischinski, Dani; Shamir, Ariel; Masia, Belen; Thies, Justus
    We introduce the Palette-Adapter, a novel method for conditioning text-to-image diffusion models on a user-specified color palette. While palettes are a compact and intuitive tool widely used in creative workflows, they introduce significant ambiguity and instability when used for conditioning image generation. Our approach addresses this challenge by interpreting palettes as sparse histograms and introducing two scalar control parameters: histogram entropy and palette-to-histogram distance, which allow flexible control over the degree of palette adherence and color variation. We further introduce a negative histogram mechanism that allows users to suppress specific undesired hues, improving adherence to the intended palette under the standard classifier-free guidance mechanism. To ensure broad generalization across the color space, we train on a carefully curated dataset with balanced coverage of rare and common colors. Our method enables stable, semantically coherent generation across a wide range of palettes and prompts. We evaluate our method qualitatively, quantitatively, and through a human evaluation, and show that it consistently outperforms existing approaches in achieving both strong palette adherence and high image quality.
  • Item
    ProcTex: Consistent and Interactive Text-to-texture Synthesis for Part-based Procedural Models
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Xu, Ruiqi; Zhu, Zihan; Ahlbrand, Benjamin; Sridhar, Srinath; Ritchie, Daniel; Masia, Belen; Thies, Justus
    Recent advances in generative modeling have driven significant progress in text-guided texture synthesis. However, current methods focus on synthesizing texture for single static 3D object, and struggle to handle entire families of shapes, such as those produced by procedural programs. Applying existing methods naively to each procedural shape is too slow to support exploring different parameter configurations at interactive rates, and also results in inconsistent textures across the procedural shapes. To this end, we introduce ProcTex, the first text-to-texture system designed for part-based procedural models. ProcTex enables consistent and real-time text-guided texture synthesis for families of shapes, which integrates seamlessly with the interactive design flow of procedural modeling. To ensure consistency, our core approach is to synthesize texture for a template shape from the procedural model, followed by a texture transfer stage to apply the texture to other procedural shapes via solving dense correspondence. To ensure interactiveness, we propose a novel correspondence network and show that dense correspondence can be effectively learned by a neural network for procedural models. We also develop several techniques, including a retexturing pipeline to support structural variation from procedural parameters, and part-level UV texture map generation for local appearance editing. Extensive experiments on a diverse set of procedural models validate ProcTex's ability to produce high-quality, visually consistent textures while supporting interactive applications. Code and data are available at: https://github.com/ruiqixu37/ProcTex.git
  • Item
    Progressively Projected Newton's Method
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Fernández-Fernández, José Antonio; Löschner, Fabian; Bender, Jan; Masia, Belen; Thies, Justus
    Newton's Method is widely used to find the solution of complex non-linear simulation problems. To guarantee a descent direction, it is common practice to clamp the negative eigenvalues of each element Hessian prior to assembly, a strategy known as Projected Newton (PN), but this perturbation often hinders convergence. In this work, we observe that projecting only a small subset of element Hessians is sufficient to secure a descent direction. Building on this insight, we introduce Progressively Projected Newton (PPN), a novel variant of Newton's Method that uses the current iterate's residual to cheaply determine the subset of element Hessians to project. The benefit is twofold: most eigendecompositions are avoided and the global Hessian remains closer to its original form, reducing the number of Newton iterations. We compare PPN with PN and Project-on-Demand Newton (PDN) in a comprehensive set of experiments covering contact-free and contact-rich deformables, co-dimensional and rigid-body simulations, and a range of time step sizes, tolerances, and resolutions. PPN reduces the amount of element projections in dynamic simulations by one order of magnitude while simultaneously improving convergence, consistently being the fastest solver in our benchmark.
  • Item
    A Discrete Polydisperse Anisotropic BSDF Model based on the Micrograin Framework
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Xu, Kewei; Lucas, Simon; Ribardière, Mickael; Bringier, Benjamin; Barla, Pascal; Masia, Belen; Thies, Justus
    We introduce a discrete polydisperse micrograin BSDF model for the rendering of porous surface materials composed of microscopic elements of different size, shape and reflectance distributed on a bulk medium. Our approach generalizes the anisotropic monodisperse micrograin model. We first reformulate it in a non-axis-aligned configuration, allowing for the later combination of different micrograin types elongated in arbitrary directions. We then extend the monodisperse model to the polydisperse case, deriving its three key components: (i) a general filling factor that controls the mix between micrograins and the bulk medium; (ii) an exact normal distribution function for surfaces composed of polydisperse micrograin distributions; and (iii) the corresponding fully-correlated shadowing and masking term. This results in an analytical single-scattering BSDF for discrete polydisperse surface materials, validated over ground truth simulations, for which we also derive a dedicated importance sampling procedure. Our model supports varying heights and anisotropy orientations of different micrograin types as input, giving additional control to simulate phenomena like retro-reflection from mixed materials, color mixture depending on lighting and observation directions, multiple directions of anisotropy, etc.
  • Item
    Skeletal-Driven Animation of Anatomical Humans via Neural Deformation Gradients
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Nolte, Gerrit; Kemper, Fabian; Schwanecke, Ulrich; Botsch, Mario; Masia, Belen; Thies, Justus
    Most real-time animation techniques for digital humans are limited to deforming the outer skin surface. Geometric skinning methods are highly efficient but struggle with artifacts such as collapsing joints or self-intersections when animating inner anatomy along with the outer skin. Volumetric physics-based simulations, on the other hand, naturally resolve these issues by coordinating bones, muscles, and skin, but are far too slow for interactive use. We solve this problem by training a neural network to predict deformation gradients. Learning deformation gradients instead of vertex displacements makes our method naturally robust to artifacts such as element inversion or volume deviation. Our model, trained on high-quality finite element simulations, generalizes well across diverse body shapes and poses. This enables anatomically consistent and physically grounded animation of bones, muscles, and skin at interactive frame rates.
  • Item
    Terrain Synthesis and Authoring based on Iso-Contours
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Huftier, Benoit; Schott, Hugo; Galin, Eric; Argudo, Oscar; Peytavie, Adrien; Guérin, Eric; Masia, Belen; Thies, Justus
    Digital terrains are central to realistic landscape depiction, yet authoring tools must balance perceptual realism with intuitive artistic control. We propose a compact vector-based representation that models terrain as nested iso-contours, inspired by geomorphology and cartography. Our method departs from traditional grid-based elevation models by generating contours through an inward Open Eden Growth simulation, followed by marching-triangles reconstruction into a Triangulated Irregular Network. This contour framework supports direct editing such as warping, slope modulation, and smoothing, while allowing reconstruction of a standard elevation map for downstream processing, including erosion and amplification. The approach enables the creation of diverse, realistic terrains from minimal user input and offers simple yet powerful control for designers.
  • Item
    Stochastic Pairwise MIS for Unbiased Large-Kernel Reuse in Real-Time
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Hedstrom, Trevor; Kettunen, Markus; Lin, Daqi; Wyman, Chris; Li, Tzu-Mao; Masia, Belen; Thies, Justus
    Spatiotemporal resampling methods such as ReSTIR decrease noise in Monte Carlo rendering of dynamic content by reusing paths across frames and pixels. Standard ReSTIR reuses spatially from a small number of randomly selected neighbors. This reuse suffers when few neighbors contain contributing samples, reducing quality toward that of the underlying path sampler. This commonly occurs during camera or object motion, as regions not present in prior frames are revealed. Increasing the number of spatial neighbors helps but also increases cost. We propose a novel spatial neighbor selection technique, stochastic pairwise MIS, which enables unbiased reuse from many neighbors in real time and focuses reuse on pixels with contributing samples. This provides a significant increase in image quality overall, especially in regions with poor input samples.
  • Item
    STAGED: Stress-Tensor Assisted Global-local-global solver for interactive Elastic shape Design
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Ruan, Liangwang; Wang, Bin; Liu, Tiantian; Chen, Baoquan; Masia, Belen; Thies, Justus
    We present an efficient and scalable method for the inverse shape design problem of elastic objects, with broad applicability to diverse materials and interactive editing. The core idea is to decouple material nonlinearity from geometry optimization by introducing the Cauchy stress tensor as an auxiliary variable. We design a three-stage scheme that iteratively optimizes the stress tensors and the rest shape, with each stage being well-posed and efficiently-solvable. To address the lack of a theoretical convergence guarantee arising from the decoupled energy formulation, we incorporate a relaxation method that ensures robust stability in practice. As a result, our method achieves a 3× speedup over the state-of-the-art asymptotic method [Jia21] on a model with 40k vertices and 112k elements, and exhibits near-linear scalability to large systems. We demonstrate applications including rest shape design for various materials, interactive material and force editing, and elastic object reconstruction from images.
  • Item
    Register-Efficient Linear-Time Evaluation in the Bernstein Basis
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Valasek, Gábor; Horváth, Anna Lili; Masia, Belen; Thies, Justus
    We investigate the evaluation of points and derivatives of Bézier curves and surfaces on modern architectures, focusing on performance and guided by numerical error bounds. While the de Casteljau algorithm remains the reference for numerical robustness, its linear working-set size imposes substantial register pressure on GPUs. We introduce a linear-time, constant-storage evaluation framework derived from the ladder algorithm that attains de Casteljau-level robustness and demonstrate that it outperforms other methods both on the GPU and CPU. Our analysis provides backward-error bounds for points and derivatives and it is also supported by empirical tests across degrees commonly used in rendering of curves and surfaces. Moreover, we show that fused multiply-add (FMA) instructions, now ubiquitous in hardware, can improve robustness even for linear interpolation. We advocate a nested FMA formulation that reconstructs endpoints exactly, in contrast to the subtraction-and-FMA pattern prevalent in shader compilers. Together, these results yield reduced memory bandwidth and register pressure, and improved performance.
  • Item
    Hierarchical Optimization of the As-Rigid-As-Possible Energy
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Meyer, Hendrik; Bickel, Bernd; Alexa, Marc; Masia, Belen; Thies, Justus
    The As-Rigid-As-Possible (ARAP) energy has become a versatile ingredient in various geometry processing and machine learning methods. The classic method for its minimization is a block coordinate descent, alternating between local rotation estimation and a global linear solve, which converges slowly for large problem instances. We develop and evaluate a multi-level scheme targeted specifically at the optimization of the ARAP energy on large meshes. The main points of our approach are (1) a mesh hierarchy that provides the necessary control over topology while being fast, (2) methods for upsampling the rotations from coarser to finer levels of the hierarchy, and (3) using direct solvers for the linear system. The resulting optimization yields smaller energy while typically being faster on a large number of test cases. The hierarchical approach generalizes to related energies and compares favorably to acceleration schemes such as ADMM, which also benefit from the hierarchical approach.
  • Item
    Latent Diffusion-GAN: Adversarial Learning in the Autoencoded Latent Space
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Jun, U-Chae; Ko, Jaeeun; Kang, Jiwoo; Masia, Belen; Thies, Justus
    Diffusion models are powerful generative frameworks for producing high-quality images by denoising latent variables from random noise. However, training with likelihood-based objectives can lead to oversmoothed high-frequency details such as textures and sharp edges. Adversarial training with GANs enhances sharpness but usually requires additional discriminator networks. We propose Latent Diffusion Generative Adversarial Networks (LD-GAN), a framework that integrates adversarial learning into diffusion models without modifying their pipeline. LD-GAN leverages the pretrained variational autoencoder as an energy-based discriminator, enabling adversarial training without extra parameters while preserving the latent priors learned from large datasets. We also introduce a structural consistency energy aligning encoder and decoder representations, improving perceptual quality. Experiments show improved sample fidelity, sharpness, and diversity across multiple generation tasks while maintaining efficient training dynamics.
  • Item
    Convex Primitive Decomposition for Collision Detection
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Knodt, Julian; Gao, Xifeng; Masia, Belen; Thies, Justus
    Creation of collision objects for 3D models is a time-consuming task requiring modelers to manually place primitives such as bounding boxes, capsules, spheres, and other convex primitives to approximate complex meshes. While automatic convex decompositions using convex hulls exist, they are often impractical for performance-sensitive applications such as games. We propose a bottom-up approach that decomposes meshes into convex primitives designed specifically for rigid-body simulation, inspired by quadric mesh simplification. Our method fits primitives to complex meshes while guaranteeing enclosure of the original surface. Experiments on over 60 models from Sketchfab demonstrate that our approach achieves lower one-way mean and median Hausdorff and Chamfer distances compared to V-HACD and CoACD while requiring less than one-third of the collider complexity. Additionally, rigid-body simulation performance measured by wall-clock time improves consistently across tested models.
  • Item
    Conversational Gesture Model (CGM): Extending Speaker-Centric Audio-Driven Motion Generation to Full Conversation Gestures
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Koren, Tomer; Rosenthal, Adi; Friedman, Doron; Shamir, Ariel; Masia, Belen; Thies, Justus
    In this work we extend speaker-centric audio-driven gesture synthesis toward a unified conversational model that jointly captures both speaking and listening behaviors. Existing speaker-centric models effectively generate gestures aligned with speech but overlook the bidirectional dynamics that characterize natural dialogue. To address this limitation, we propose the Conversational Gesture Model (CGM), a cross-attention-based model capable of synthesizing gestures conditioned on interlocutor conversational cues such as gestures, tone, and textual semantics. By leveraging cross-attention mechanisms, the model fuses interlocutor audio and text features with character gesture encodings, enabling a single system to seamlessly alternate between speaking and listening roles of the same character. Hence, our model enables a single system to act as both speaker and listener, capturing the fluid role shifts and mutual influence inherent in conversation. Experiments demonstrate that this approach preserves the quality of speaker-driven gestures while significantly improving the realism, coherence, and responsiveness of full conversational interactions.
  • Item
    Mixed Super-Circles
    (The Eurographics Association and John Wiley & Sons Ltd., 2026) Hohnadel, Emile; Métivet, Thibaut; Bertails-Descoubes, Florence; Masia, Belen; Thies, Justus
    We introduce mixed super-circles, a position-curvature formulation of the original dynamic 2D super-helix model. Compared to the purely curvature-based chained formulation, the mixed formulation drastically reduces the algorithmic complexity of the solving scheme from quadratic to quasi-linear and simplifies the handling of positional constraints including contacts. It recovers the advantages of classical position-based models while preserving the high-order convergence of curvature-based models. The smooth piecewise circular arc representation avoids spurious jumps in contact forces that are difficult to eliminate with position-based models. The model is validated quantitatively against demanding mechanical tests involving contact, friction, snapping and vibrations. Its versatility, robustness and efficiency are demonstrated through interactive scenarios with multiple planar elastic rods under various boundary conditions and constraints. The corresponding source code, Circonflex, is released under the GNU GPL v3 licence.