SIGGRAPH Asia 2025 Technical Papers
318 papers in 54 sessions, organized according to the official SIGGRAPH Asia 2025 program. Paper titles link to ACM Digital Library. Awards follow the official awards announcement.
Best Paper Honorable Mention
Monday, 15 December 2025
3D Reconstruction & Intelligent Geometry
- RCTrans: Transparent Object Reconstruction in Natural Scene via Refractive Correspondence Estimation
- Topology-Aware Optimization of Gaussian Primitives for Human-Centric Volumetric Videos
- Surface-Aware Distilled 3D Semantic Features
- Lang3D-XL: Language Embedded 3D Gaussians for Large-scale Scenes
- PoissonNet: A Local-Global Approach for Learning on Surfaces
- PhysFiT: Physical-aware 3D Shape Understanding for Finishing Incomplete Assembly
Dynamic Generative Video: From Synthesis To Real-Time Editing
- Navigating with Annealing Guidance Scale in Diffusion Space
- Input-Aware Sparse Attention for Real-Time Co-Speech Video Generation
- OmnimatteZero: Fast Training-free Omnimatte with Pre-trained Video Diffusion Models
- STGlight: Online Indoor Lighting Estimation via Spatio-Temporal Gaussian Fusion
- GenLit: Reformulating Single-Image Relighting as Video Generation
- DynVFX: Augmenting Real Videos with Dynamic Content
Global Illumination & Real-Time Rendering
- Vertex Features for Neural Global Illumination
- NeLiF: Neural Lighting Function Generation for Real-Time Indoor Rendering
- ReSTIR PG: Path Guiding with Spatiotemporally Resampled Paths
- Sample Space Partitioning and Spatiotemporal Resampling for Specular Manifold Sampling
- GS-ROR2: Bidirectional-guided 3DGS and SDF for Reflective Object Relighting and Reconstruction
- Frame-Free Representation of Polarized Light for Resolving Stokes Vector Singularities
High-Performance Simulation Algorithms
- Viscous Vortex Dynamics on Surfaces
- Fast Galerkin Multigrid Method for Unstructured Meshes
- A Stack-Free Parallel h-Adaptation Algorithm for Dynamically Balanced Trees on GPUs
- Implicit Position Based Fluids
- Reliable Iterative Dynamics: A Versatile Method for Fast and Robust Simulation
- Implicit Bonded Discrete Element Method with Manifold Optimization
Camera Control And Directed Storytelling In Video Generation
- Shape-for-Motion: Precise and Consistent Video Editing With 3D Proxy
- Context as Memory: Scene-Consistent Interactive Long Video Generation with Memory Retrieval
- CamCloneMaster: Enabling Reference-based Camera Control for Video Generation
- Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generation
- Cut2Next: Generating Next Shot via In-Context Tuning
- CamPVG: Camera-Controlled Panoramic Video Generation with Epipolar-Aware Diffusion
Material & Texture Modeling
- Scattering-Aware Color Calibration for 3D Printers Using a Simple Calibration TargetHonorable Mention
- Example-Based Feature Painting on Textures
- DiffTex: Differentiable Texturing for Architectural Proxy Models
- Fine-Grained Spatially Varying Material Selection in Images
- S3 Imagery: Specular Shading from Scratch-Anisotropy
- SeqTex: Generate Mesh Textures in Video Sequence
Mesh Processing
- Simplifying Textured Triangle Meshes in the Wild
- PartUV: Part-Based UV Unwrapping of 3D Meshes
- Local Surface Parameterizations via Smoothed Geodesic Splines
- Solid-Shell Labeling for Discrete Surfaces
- RibbonSculpt: Voronoi Ball based 3D Sculpting from Sparse VR Ribbons
- A fast, efficient, and robust feature protected denoising method
Neural & Implicit Representations For Geometry And Physics
- Precise Gradient Discontinuities in Neural Fields for Subspace Physics
- Variational Neural Surfacing of 3D Sketches
- NeuVAS: Neural Implicit Surfaces for Variational Shape Modeling
- Neural Kinematic Bases for Fluids
- Neural Octahedral Field: Octahedral Prior for Simultaneous Smoothing and Sharp Edge Regularization
- Practical Gaussian Process Implicit Surfaces with Sparse Convolutions
Creating Digital Humans
- Generative Head-Mounted Camera Captures for Photorealistic Avatars
- InfiniHuman: Realistic 3D Human Creation with Precise Control
- HRM^2Avatar: High-Fidelity Real-Time Mobile Avatars from Monocular Phone Scans
- PriorAvatar: Efficient and Robust Avatar Creation from Monocular Video Using Learned Priors
- Constructing Diffusion Avatar with Learnable Embeddings
- HumanLift: Single-Image 3D Human Reconstruction with 3D-Aware Diffusion Priors and Facial Enhancement
Physically Based Simulation & Dynamic Environments
- A Highly-Efficient Hybrid Simulation System for Flight Controller Design and Evaluation of Unmanned Aerial VehiclesHonorable Mention
- PhySIC: Physically Plausible 3D Human-Scene Interaction and Contact from a Single Image
- FreeMusco: Motion-Free Learning of Latent Control for Morphology-Adaptive Locomotion in Musculoskeletal Characters
- CFC: Simulating Character-Fluid Coupling using a Two-Level World Model
- Fast & Stable Control of Coupled Solid-Fluid Dynamic Systems
- Force-Dual Modes: Subspace Design from Stochastic Forces
Smart Process Planning For Manufacturing
- Waste-to-Value: Reutilized Material Maximization for Additive and Subtractive Hybrid Remanufacturing
- Can Any Model Be Fabricated? Inverse Operation Based Planning for Hybrid Additive–Subtractive Manufacturing
- Chapper: Carvable Hull-and-Pack for Subtractive Manufacturing
- MiGumi: Making Tightly Coupled Integral Joints Millable
- Curve-Based Slicer for Multi-Axis DLP 3D PrintingBest Paper
- INF-3DP: Implicit Neural Fields for Collision-Free Multi-Axis 3D Printing
Visibility & Real-Time Rendering
- NeuralPVS: Learned Estimation of Potentially Visible Sets
- Neural Visibility of Point Sets
- Potentially Visible Set Generation with the Disocclusion Buffer
- Lightweight, Edge-Aware, and Temporally Consistent Supersampling for Mobile Real-Time Rendering
- Consecutive Frame Extrapolation with Predictive Sparse Shading
- PatchEX: High-Quality Real-Time Temporal Supersampling through Patch-based Parallel Extrapolation
Tuesday, 16 December 2025
Audio-Driven Facial And Portrait Animation
- Audio Driven Real-Time Facial Animation for Social Telepresence
- LSF-Animation: Label-Free Speech-Driven Facial Animation via Implicit Feature Representation
- High-Fidelity Dynamic Portrait Animation via Direct Preference Optimization and Temporal Motion Modulation
- Audio-Driven Universal Gaussian Head Avatars
- ARTalk: Speech-Driven 3D Head Animation via Autoregressive Model
- X-Actor: Emotional and Expressive Long-Range Portrait Acting from Audio
Computational Design & Fabricability
- LEGO-Maker: Autoregressive Image-Conditioned LEGO Model Creation
- LegoACE: Autoregressive Construction Engine for Expressive LEGO Assemblies
- Computational Design of Shape-Aware Sieves
- Designing with Tension: Nearly-Developable Patch Layouts
- Inverse Tiling of 2D Finite Domains
- Fabricable Discretized Ruled Surfaces
Computational Photography & Cameras
- CameraVDP: Perceptual Display Assessment with Uncertainty Estimation via Camera and Visual Difference Prediction
- DiffCamera: Arbitrary Refocusing on Images
- Learning to Refocus with Video Diffusion Models
- Generating the Past, Present and Future from a Motion-Blurred Image
- Automated Design of Compound Lenses with Discrete-Continuous Optimization
- UltraZoom: Generating Gigapixel Images from Regular Photos
Sampling, Reconstruction & Variance Reduction
- Nonlinear Noise2Noise for Efficient Monte Carlo Denoiser Training
- Statistical Error Reduction for Monte Carlo Rendering
- DSCombiner: Double Shrinkage for Combining Biased and Unbiased Monte Carlo Renderings
- Jackknife Transmittance and MIS Weight Estimation
- Imperfect Image-Space Control Variates for Monte Carlo Rendering
- SZ Sequences: Binary-Constructed $(0, 2^q)$-Sequences
Differentiable Rendering & Applications
- Automatic Sampling for Discontinuities in Differentiable ShadersBest Paper
- Many-Worlds Inverse Rendering
- Differentiable Light Transport with Gaussian Surfels via Adapted Radiosity for Efficient Relighting and Geometry Reconstruction
- Spectral Reconstruction with Uncertainty Quantification via Differentiable Rendering and Null-Space Sampling
- Adaptive Neural Kernels for Gradient-domain Rendering
- Generalized Unbiased Reconstruction for Gradient-Domain Rendering
Generative 3D Shape Synthesis
- ShapeGen: Towards High-Quality 3D Shape Synthesis
- Autoregressive Generation of Static and Growing Trees
- OmniPart: Part-Aware 3D Generation with Semantic Decoupling and Structural Cohesion
- ART-DECO: Arbitrary Text Guidance for 3D Detailizer Construction
- SPGen: Spherical Projection as Consistent and Flexible Representation for Single Image 3D Shape Generation
Image Restoration, Editing & Enhancement
- DvD: Unleashing a Generative Paradigm for Document Dewarping via Coordinates-based Diffusion Model
- HRC-Net: Learning Visual Hypothesis, Representative, and Collaboration for Multi-Domain Image Inpainting
- Harnessing Diffusion-Yielded Score Priors for Image Restoration
- ELAD: Blind Face Restoration using Expectation-based Likelihood Approximation and Diffusion Prior
- Self-supervised Underwater Color Restoration via Wavelet-Diffusion Model with Filtered Multi-Scale Feature Distillation
- Self-supervised Texture Filtering
Perception And Performance In AR/VR Systems
- Supra-threshold Contrast Perception in Augmented Reality
- Vertical Binocular Misalignment in AR Impairs Reading Performance
- Modeling and Exploiting the Time Course of Chromatic Adaptation for Display Power Optimizations in Virtual Reality
- Performance Analysis of Catch-Up Eye Movements in Visual Tracking
- Glare Pattern Depiction: High-Fidelity Physical Computation and Physiologically-Inspired Visual Response
- MILO: A Lightweight Perceptual Quality Metric for Image and Latent-Space OptimizationHonorable Mention
3D Reconstruction & Rendering
- EGG-Fusion: Efficient 3D Reconstruction with Geometry-aware Gaussian Surfel on the Fly
- Inverse Radiative Transport for Infrared Scenes with Gaussian Primitives
- Efficient Object Reconstruction with Differentiable Area Light Shading
- WATER: Watertight Tessellation for Real-Time Pixel-Accurate Rendering of Large-Scale Surfaces
- GigaSLAM: Large-Scale Monocular SLAM with Hierarchical Gaussian Splats
- Artifact-Resilient Real-Time Holography
4D Gaussian Splatting For Dynamic Scene Reconstruction
- TrackerSplat: Exploiting Point Tracking for Fast and Robust Dynamic 3D Gaussians Reconstruction
- Clustered Error Correction with Grouped 4D Gaussian Splatting
- Anchored 4D Gaussian Splatting for Dynamic Novel View Synthesis
- Prior-Enhanced Gaussian Splatting for Dynamic Scene Reconstruction from Casual Video
- 4DSloMo: 4D Reconstruction for High Speed Scene with Asynchronous Capture
- Split4D: Decomposed 4D Scene Reconstruction Without Video Segmentation
Animation, Simulation & Deformation
- Numerical Homogenization of Sand from Grain-level Simulations
- Improving Curl Noise
- Gaussian See, Gaussian Do: Semantic 3D Motion Transfer from Multiview Video
- QMF-Blend: Quantized Matrix Factorization for Efficient Blendshape Compression
- AniMaker: Multi-Agent Animated Storytelling with MCTS-Driven Clip Generation
- Shape-aware Inertial Poser: Motion Tracking for Humans with Diverse Shapes Using Sparse Inertial Sensors
Garment & Cloth Modeling, Simulation And Rendering
- GarmageNet: A Multimodal Generative Framework for Sewing Pattern Design and Generic Garment Modeling
- Realistic Cloth Rendering with a Ray-Wave Hybrid Shading Model
- Progressive Outfit Assembly and Instantaneous Pose Transfer
- Neighbor-Aware Data-Driven Relaxation of Stitch Mesh Models for Knits
- A Nonconforming Formulation of Cloth
Intelligent CAD: B-Reps, NURBs & Splines
- AutoBrep: Autoregressive B-Rep Generation with Unified Topology and Geometry
- BrepGPT: Autoregressive B-rep Generation with Voronoi Half-Patch
- NURBS-Based Grid Shell Form Finding on Domains with Topologically Arbitrary Boundaries
- Overlap Region Extraction of Two NURBS Surfaces
- $G^2$ Interpolating Spline with Local Maximum Curvature
- Img2CAD: Reverse Engineering 3D CAD Models from Images through VLM-Assisted Conditional Factorization
It's All About The Motion
- Curvature Enthusiasm: Correspondence-Free Interpolation and Matching of Articulated 3D Shapes using Compressed Normal Cycles
- Motion In-Betweening for Densely Interacting Characters
- Control Operators for Interactive Character AnimationBest Paper
- Environment-aware Motion Matching
- StableMotion: Training Motion Cleanup Models with Unpaired Corrupted Data
- Learning to Ball: Composing Policies for Long-Horizon Basketball Moves
Neural Fields And Surface Reconstruction
- Low-Rank Adaptation of Neural Fields
- Spectral Prefiltering of Neural Fields
- Marching Neurons: Accurate Surface Extraction for Neural Implicit Shapes
- SOF: Sorted Opacity Fields for Fast Unbounded Surface Reconstruction
- MILo: Mesh-In-the-Loop Gaussian Splatting for Detailed and Efficient Surface Reconstruction
- RaRa Clipper: A Clipper for Gaussian Splatting Based on Ray Tracer and Rasterizer
Vector Graphics & Sketches
- Capturing Non-Linear Human Perspective in Line Drawings
- AutoSketch: VLM-assisted Style-Aware Vector Sketch Completion
- KISSColor: Kinetic and Intuitive Stroke Stretching for Vector Drawing Colorization
- LayerPeeler: Autoregressive Peeling for Layer-wise Image Vectorization
- Neural Image abstraction using long smoothing B-splines
- UTDesign: A Unified Framework for Stylized Text Editing and Generation in Graphic Design Images
Wednesday, 17 December 2025
Compositional And Layout-Guided Image Synthesis
- Object-level Visual Prompts for Compositional Image Generation
- ComposeMe: Attribute-Specific Image Prompts for Controllable Human Image Generation
- PartComposer: Learning and Composing Part-Level Concepts from Single-Image Examples
- AGSwap: Overcoming Category Boundaries in Object Fusion via Adaptive Group Swapping
- MALeR: Improving Compositional Fidelity in Layout-Guided Generation
Computational Design & Geometry
- One String to Pull Them All: Fast Assembly of Curved Structures from Flat Auxetic Linkages
- Discovering Folding Lines for Surface Compression
- Reconfigurable Hinged Kirigami Tessellations
- Snapping Deployable Toroids for Modular Gridshells
- Closed-Form Construction of Voronoi Diagrams with Star-Shaped Metrics
- Star-Shaped Distance Voronoi Diagrams for 3D Metamaterial Design
Hair & Faces
- Auto Hair Card Extraction for Smooth Hair with Differentiable Rendering
- Strands2Cards: Automatic Generation of Hair Cards from Strands
- Shaping Strands with Neural Style Transfer
- CHARM: Control-point-based 3D Anime Hairstyle Auto-Regressive Modeling
- Single-Shot Facial Capture using Polarized RGB Sinusoidal Illumination
- Single Image 3D Portrait Relighting with Generative Priors
Differentiable Physics And Fabrication-Aware Optimization
Generative Scenes & Panoramas
- VideoFrom3D: 3D Scene Video Generation via Complementary Image and Video Diffusion Models
- SS4D: Native 4D Generative Model via Structured Spacetime Latents
- Generating 360° Video is What You Need For a 3D Scene
- PanoDreamer: Optimization-Based Single Image to 360 3D Scene With Diffusion
- WorldExplorer: Towards Generating Fully Navigable 3D Scenes
- Voyager: Long-Range and World-Consistent Video Diffusion for Explorable 3D Scene Generation
Human & Robot Animation & Behavior
- Social Agent: Mastering Dyadic Nonverbal Behavior Generation via Conversational LLM Agents
- How Does a Virtual Agent Decide Where to Look? Symbolic Cognitive Reasoning for Embodied Head Rotation
- Echo: Enhancing Conversational Behavior Generation via Hierarchical Semantic Comprehension with Large Language Models
- SymBridge: A Human-in-the-Loop Cyber-Physical Interactive System for Adaptive Human-Robot Symbiosis
- Unifying Latent Action and Latent State Pre-training for Policy Learning from Videos
- JoruriPuppet: Learning Tempo-Changing Mechanisms Beyond the Beat for Music-to-Motion Generation with Expressive Metrics
4D & Dynamic Scene Generation And Reconstruction
- PAD3R: Pose-Aware Dynamic 3D Reconstruction from Casual Videos
- Detail Enhanced Gaussian Splatting for Large-Scale Volumetric Capture
- GS-RoadPatching: Inpainting Gaussians via 3D Searching and Placing for Driving Scenes
- Neural Hamiltonian Deformation Fields for Dynamic Scene Rendering
- MVP4D: Multi-View Portrait Video Diffusion for Animatable 4D Avatars
- MV-Performer: Taming Video Diffusion Model for Faithful and Synchronized Multi-view Performer Synthesis
Advanced Light Transport & PDE Solvers
- Jump Restore Light Transport
- Gaussian Integral Linear Operators for Precomputed GraphicsHonorable Mention
- Harmonic Caching for Walk on Spheres
- Robust Derivative Estimation with Walk on Stars
- Off-Centered WoS-Type Solvers with Statistical Weighting
- An Adjoint Method for Differentiable Fluid Simulation on Flow Maps
Efficient And Robust Algorithms For Geometric Computing
- BSP-OT: Sparse transport plans between discrete measures in loglinear timeBest Paper
- Efficient and Scalable Spatial Regularization of Optimal TransportHonorable Mention
- Medial Sphere Preconditioning for Knot Untangling and Volume-Filling Curves
- CrossGen: Learning and Generating Cross Fields for Quad Meshing
- Sums of Wedges: Conforming Weighted Delaunay Triangulations are Polynomial in Fixed Dimension
- Lifted Surfacing of Generalized Sweep Volumes
3D Reconstruction & View Synthesis
- RemixFusion: Residual-based Mixed Representation for Large-scale Online RGB-D Reconstruction
- Convection Augmented Gauss Reconstruction for Unoriented Point Clouds
- AD-GS: Alternating Densification for Sparse-Input 3D Gaussian Splatting
- Training-Free Instance-Aware 3D Scene Reconstruction and Diffusion-Based View Synthesis from Sparse Images
- MODepth: Benchmarking Mobile Multi-frame Monocular Depth Estimation with Optical Image Stabilization
Animating Images, Sketches And Text
- From Rigging to Waving: 3D-Guided Diffusion for Natural Animation of Hand-Drawn Characters
- Sketch2PoseNet: Efficient and Generalized Sketch to 3D Human Pose Prediction
- AnimaX: Animating the Inanimate in 3D with Joint Video-Pose Diffusion Models
- Animus3D: Text-driven 3D Animation via Motion Score Distillation
- X-UniMotion: Animating Human Images with Expressive, Unified and Identity-Agnostic Motion Latents
- FairyGen: Storied Cartoon Video from a Single Child-Drawn Character
Real-Time Rendering & System Optimization
- Evaluating and Sampling Glinty NDFs in Constant Time
- 3D SMoE Splatting for Edge-aware Realtime Radiance Field Rendering
- AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views
- PowerGS: Display-Rendering Power Co-Optimization for Neural Rendering in Power-Constrained XR Systems
- Sparse Cache Updates for Scalable Distributed Effect-Based Rendering
- StereoFG: Generating Stereo Frames from Centered Feature Stream
Cameras, Sensors, And Acquisition
- Fovea Stacking: Imaging with Dynamic Localized Aberration CorrectionBest Paper
- Underwater Optical Backscatter Communications using Acousto-Optic Beam Steering
- Shoot-Bounce-3D: Single-Shot Occlusion-Aware 3D from Lidar by Decomposing Two-Bounce Light
- Robust Single-shot Structured Light 3D Imaging via Neural Feature Decoding
- Transient LASSO: Transient Large-Scale Scene Reconstruction
- LookUp3D: Data-Driven 3D Scanning
Generative 3D Modeling
- Imaginarium: Vision-guided High-Quality 3D Scene Layout Generation
- Procedural Scene Programs for Open-Universe Scene Generation: LLM-Free Error Correction via Program Search
- Hierarchical Neural Semantic Representation for 3D Semantic Correspondence
- StyleSculptor: Zero-Shot Style-Controllable 3D Asset Generation with Texture-Geometry Dual Guidance
- Fuse3D: Generating 3D Assets Controlled by Multi-Image Fusion
- In-2-4D: Inbetweening from Two Single-View Images to 4D Generation
Motion Transfer & Control
- Motion2Motion: Cross-topology Motion Retargeting with Sparse Correspondence
- Ultrafast and Controllable Online Motion Retargeting for Game Scenarios
- SMF: Template-free and Rig-free Animation Transfer using Kinetic Codes
- PhysHMR: Learning Humanoid Control Policies from Vision for Physically Plausible Human Motion Reconstruction
- Generating Detailed Character Motion from Blocking Poses
- MaskedManipulator: Versatile Whole-Body Manipulation
Thursday, 18 December 2025
Advanced Fluid And Multiphase Simulation
- The Granule-In-Cell Method for Simulating Sand--Water Mixtures
- Kinetic Free-Surface Flows and Foams with Sharp Interfaces
- Implicit Incompressible Porous Flow using SPH
- Fire-X: Extinguishing Fire with Stoichiometric Heat Release
- Multiphase Particle-Based Simulation of Poro-Elasto-Capillary Effects
- Wavelet Fluids
Material & Reflectance Modeling
- Diffusion-Guided Relighting for Single-Image SVBRDF Estimation
- EBREnv: SVBRDF Estimation in Uncontrolled Environment Lighting via Exemplar-Based Representation
- Sparse SVBRDF Acquisition via Importance-Aware Illumination Multiplexing
- Chord: Chain of Rendering Decomposition for PBR Material Estimation from Generated Texture Images
- AniTex: Light-Geometry Consistent PBR Material Generation for Animatable Objects
- Hyperspectral Polarimetric BRDFs of Real-world Materials
Objects In Parts & Articulation
- FreeArt3D: Training-Free Articulated Object Generation using 3D Diffusion
- Generating Objects with Part-Articulation from a Single Image
- LARM: A Large Articulated Object Reconstruction Model
- ArtiLatent: Realistic Articulated 3D Object Generation via Structured Latents
- Assembler: Scalable 3D Part Assembly via Anchor Point Diffusion
- LLM-Primitives: Large Language Model for 3D Reconstruction with Primitives
Text-To-Image & Customization
- ConsiStyle: Style Diversity in Training-Free Consistent T2I Generation
- Bokeh Diffusion: Defocus Blur Control in Text-to-Image Diffusion Models
- PractiLight: Practical Light Control Using Foundational Diffusion Models
- Magic Fixup: Streamlining Photo Editing by Watching Dynamic Videos
- Zero-Shot Dynamic Concept Personalization with Grid-Based LoRA
- B4M: Breaking Low-Rank Adapter for Making Content-Style Customization
Expressive And Structured Gaussian Representations
- Gradient-Weighted Feature Back-Projection: A Fast Alternative to Feature Distillation in 3D Gaussian Splatting
- Rigidity-Aware 3D Gaussian Deformation from a Single Image
- Neural Texture Splatting: Expressive 3D Gaussian Splatting for View Synthesis, Geometry, and Dynamic Reconstruction
- JumpingGS: Level-jump 3D Gaussian Representation for Delicate Textures in Aerial Large-scale Scene Rendering
- TC-GS: A Faster Gaussian Splatting Module Utilizing Tensor Cores
- DeMapGS: Simultaneous Mesh Deformation and Surface Attribute Mapping via Gaussian Splatting
Generative Synthesis, Editing & Customization
- One-shot Embroidery Customization via Contrastive LoRA Modulation
- Teamwork: Collaborative Diffusion with Low-rank Coordination and Adaptation
- Virtually Being: Customizing Camera-Controllable Video Diffusion Models with Volumetric Performance Captures
- Proteus-ID: ID-Consistent and Motion-Coherent Video Customization
- DreamID: High-Fidelity and Fast diffusion-based Face Swapping via Triplet ID Group Learning
- HiWave: Training-Free High-Resolution Image Generation via Wavelet-Based Diffusion Sampling
Human Motion Synthesis & Interaction
- Physics-Based Motion Imitation with Adversarial Differential Discriminators
- Learning Human Motion with Temporally Conditional Mamba
- SRBTrack: Terrain-Adaptive Tracking of a Single-Rigid-Body Character Using Momentum-Mapped Space-Time Optimization
- Uni-Inter Unifying 3D Human Motion Synthesis Across Diverse Interaction Contexts
- HOMA: Towards Generic Human-Object Interaction in Multimodal Driven Human Animation with Weak Conditions
- CHOICE: Coordinated Human-Object Interaction in Cluttered Environments for Pick-and-Place Actions
Shape Abstraction And Structural Analysis
- RL-ACD: Reinforcement Learning-based Approximate Convex Decomposition
- ASIA: Adaptive 3D Segmentation using Few Image Annotations
- Light-SQ: Structure-aware Shape Abstraction with Superquadrics for Generated Meshes
- MATStruct: High-quality Medial Mesh Computation via Structure-aware Variational Optimization
- Temporally Smooth Mesh Extraction for Procedural Scenes with Long-Range Camera Trajectories using Spacetime Octrees
- Design for Descent: What Makes a Shape Grammar Easy to Optimize?
Advanced Representations And Rendering For 3D Scenes
- LVT: Large-Scale Scene Reconstruction via Local View Transformers
- CityGo: Lightweight Urban Modeling and Rendering with Proxy Buildings and Residual Gaussians
- GSWT: Gaussian Splatting Wang Tiles
- Spectral-GS: Taming 3D Gaussian Splatting with Spectral Entropy
- A compact stochastic representation for Monte Carlo Path Traced images
- Editable Physically-based Reflections in Raytraced Gaussian Radiance Fields
Diffusion-Based Image Editing & Manipulation
- In-Context Brush: Zero-shot Customized Subject Insertion with Context-Aware Latent Space Manipulation
- BlobCtrl: Taming Controllable Blob for Element-level Image Editing
- ConsistEdit: Highly Consistent and Precise Training-free Visual Editing
- Voost: A Unified and Scalable Diffusion Transformer for Bidirectional Virtual Try-On and Try-Off
- DreamO: A Unified Framework for Image Customization
- The Aging Multiverse: Generating Condition-Aware Facial Aging Tree via Training-Free Diffusion
Geometry Processing & Representations
- Closed-form Cauchy Coordinates and Their Derivatives for 2D High-order Cages
- Aerial Path Planning for Urban Geometry and Texture Co-Capture
- Direct Rendering of Intrinsic Triangulations
- Compact shape representation utilizing local surface similarities
- NESI: Neural Explicit-Shape-Intersection-Based Geometry Representation