Yebin Liu

AnimatableGaussians source code released! github.com/lizhe00/Animatabl…

We have build a code repository of our group projects and made public some source codes, including the latest works of DreamCraft3D, GPS-Gaussian, HumanNorm, and HAvatar. See: github.com/thu3dhuman

Gaussian Splatting relies on per-frame parameter optimizations for several minutes. Our generalizable pixel-wise 3D Gaussian Splatting enable instantly and realtime rendering without necessitating any fine-tuning or optimization for an unseen character. shunyuanzheng.github.io/GPS-…

Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling Projectpage: animatable-gaussians.github.…

Our new Siggaph Asia 2022 paper proposes a real-time 3D-aware face editing framework that supports high-fidelity 3D face style editing. mrtornado24.github.io/IDE-3D…

Our new Siggaph Asia 2022 paper proposes a real-time 3D-aware face editing framework that supports high-fidelity facial synthesis and fine-grained style editing of shape and appearance. mrtornado24.github.io/IDE-3D…

Source code (including training code) public at: github.com/jsnln/fite #ECCV2022 Learning Implicit Templates for Point-Based Clothed Human Modeling jsnln.github.io/fite/

#ECCV2022 (Oral) Diffusion model is also good at stereo matching! We show high quality human reconstruction using only 8 cameras, by learning diffuse stereo matching on human scan datasets. Video demo uploaded on webpage: liuyebin.com/diffustereo/dif…

#ECCV2022(Oral). Diffusion model is also good at stereo matching! In this work, by learning diffusion stereo matching on the high quality human scan datasets, we achieve high quality reconstruction using only 8 camera views on real world data. Project: liuyebin.com/diffustereo/dif…

#ECCV2022. Diffusion model is also good at stereo matching! By learning diffusion stereo matching on the human scan datasets, we achieve extremely high quality reconstruction using only 8 RGB views. Results on synthetic dataset: Project: liuyebin.com/diffustereo/dif…

#ECCV2022 This works solves two challenges in e Single-image Full-body Human Relighting: (i) inability to disambiguate albedo and lighting (ii) inability to model high-frequency shadows due to reliance on Spherical harmonics representation of lighting. jcn16.github.io/jcn.github.i…

#ECCV2022 FITE: A First-Implicit-Then-Explicit framework for modeling humans in clothing. Leveraging both implicit and explicit representations, FITE produces more expressive clothing details and achieves better geometric continuity. project page: jsnln.github.io/fite/

PyMAF-X: Towards Well-aligned Full-body Model Regression from Monocular Images arxiv:arxiv.org/abs/2207.06400 project:hongwenzhang.github.io/pymaf…

THuman4.0 Dataset contains multi-view video sequences captured with 24 well-calibrated RGB cameras in 30 fps, with a resolution of 1330x1150. We also provide the SMPL fitting in the dataset. Page: github.com/ZhengZerong/THUma…

FaceVerse: a deep Morphable model for chinese faces using hybrid datasets. CVPR Poster 4.2 24-Jun-22 Representation Learning 2:30 PM 5:00 PM code and dataset: liuyebin.com/faceverse/facev…

DoubleField for High-fidelity Human Reconstruction and Rendering. Using only a few RGB cameras (<5) for instant reconstruction and rendering of human actors. CVPR Poster 3.2 23-Jun Video Analysis & Understanding 2:30 PM 5:00 PM. liuyebin.com/dbfield/dbfield…