Official PyTorch implementation of "Expressive Whole-Body 3D Gaussian Avatar", ECCV 2024.

Vid2Avatar: 3D Avatar Reconstruction from Videos in the Wild via Self-supervised Scene Decomposition (CVPR2023)

Official implementation of "En3D: An Enhanced Generative Model for Sculpting 3D Humans from 2D Synthetic Data", CVPR 2024; 3D Avatar Generation and Animation

Implementation of "Image Restoration Through Generalized Ornstein-Uhlenbeck Bridge", accepted by ICML 2024.

Image Restoration with Mean-Reverting Stochastic Differential Equations, ICML 2023. Winning solution of the NTIRE 2023 Image Shadow Removal Challenge.

[CVPR 2024] Structure Matters: Tackling the Semantic Discrepancy in Diffusion Models for Image Inpainting

[CVPR2024 (Highlight)] RichDreamer: A Generalizable Normal-Depth Diffusion Model for Detail Richness in Text-to-3D. Live Demo：https://modelscope.cn/studios/Damo_XR_Lab/3D_AIGC

GPU Accelerated Non-rigid ICP for surface registration

Implicit Feature Network (IF-Net) - Codebase

Implicit Feature Network (IF-Net) for Texture Completion - Codebase.

ReLoo: Reconstructing Humans Dressed in Loose Garments from Monocular Video in the Wild (ECCV2024)

SMPLitex – Official implementation

[CVPR'23, Highlight] ECON: Explicit Clothed humans Optimized via Normal integration

[NeurIPS 2022] Non-rigid Point Cloud Registration with Neural Deformation Pyramid

Open3D-based implementation of DynamicFusion (2015), python implementation of NeurIPS 2020 paper Neural Non-Rigid Tracking,.

[SIGGRAPH Asia 2024 (Journal Track)] StableNormal: Reducing Diffusion Variance for Stable and Sharp Normal

StableDelight: Revealing Hidden Textures by Removing Specular Reflections

Three.js-based implementation of 3D Gaussian splatting

More relighting!

[NeurIPS 2024] GaussianCube: A Structured and Explicit Radiance Representation for 3D Generative Modeling

[CVPR2024] Official implementation of SplattingAvatar.

Continuous Remeshing For Inverse Rendering

[NeurIPS 2024] Unique3D: High-Quality and Efficient 3D Mesh Generation from a Single Image

Offical codes for "GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-Mesh"

🌊 [ECCV'24 Oral] MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images

Official implementation of "Bilateral Normal Integration" (BiNI), ECCV 2022.

[ECCV'24] GeoWizard: Unleashing the Diffusion Priors for 3D Geometry Estimation from a Single Image