Home   ·   Publications   ·   More
I am a Ph.D. student at MMLab@NTU, Nanyang Technological University,
supervised by Prof. Ziwei Liu.
My research focuses on 3D human modeling. I am particularly interested in the perception and synthesis of human motion and human-environment interactions.
|
![]() |
[2022-07]   One paper (HuMMan) accepted to ECCV 2022 for Oral presentation.
[2022-05]   One paper (AvatarCLIP) accepted to SIGGRAPH 2022 (journal track).
[2022-03]   Two papers, Balanced MSE (Oral) and GE-ViTs, accepted to CVPR 2022.
[2022-01]   One paper (BiBERT) accepted to ICLR 2022.
[2021-08]   Start my journey at MMLab@NTU!
* indicates equal contribution, ✉ indicates corresponding / co-corresponding author
![]() |
arXiV, 2022 [Paper]  [Project Page]  [Video] [Code] [Colab Demo] [Hugging Face Demo] Star The first text-driven motion generation pipeline based on diffusion models with probabilistic mapping, realistic synthesis and multi-level manipulation ability.
|
![]() |
ACM Transactions on Graphics (SIGGRAPH), 2022 [Paper]  [Project Page]  [Video] [Code] [Colab Demo] Star AvatarCLIP is the first zero-shot text-driven pipeline, which empowers layman users to generate and animate 3D Avatars by natural language description.
|
![]() |
Conference on Computer Vision and Pattern Recognition (CVPR), 2022 (Oral Presentation) [Paper]  [Project Page]  [Talk] [Code] [Hugging Face Demo] Star A statistically principled loss function to address the train/test mismatch in imbalanced regression, coincides with the supervised contrastive loss.
|
![]() |
Conference on Computer Vision and Pattern Recognition (CVPR), 2022 [Paper]  [Code] Star A systematical comparison of the generalization ability between CNNs and ViTs. Three representative generalization-enhancement techniques are applied to ViTs to further explore their inner properties.
|
![]() |
Zhongang Cai*,
Mingyuan Zhang*,
Jiawei Ren*,
Chen Wei,
Daxuan Ren,
Zhengyu Lin,
Haiyu Zhao,
Lei Yang,
Chen Change Loy,
Ziwei Liu✉
arXiV, 2021 [Paper]  [Code] Star A large-scale synthetic human dataset collected using GTA-5 game engine, providing stable performance boost to both frame-based and video-based HMR.
|
![]() |
International Conference on Learning Representations (ICLR), 2022 [Paper]  [Code] Star BiBERT is the first fully binarized BERT. It introduces an efficient Bi-Attention structure and a DMD scheme, which yields impressive 59.2x and 31.2x saving on FLOPs and model size.
|
![]() |
International Conference on Learning Representations (ICLR), 2021 [Paper]  [Code] Star BiPointNet is the first fully binarized network for point cloud learning. BiPointNet gives an impressive 14.7x speedup and 18.9x storage saving on real-world resource-constrained devices.
|
![]() |
Winter Conference on Applications of Computer Vision (WACV), 2021 [Paper]  [Code] Star Efficient Attention reduces the memory and computational complexities of the attention mechanism from quadratic to linear. It demonstrates significant improvement in performance-cost trade-offs on a variety of tasks including object detection, instance segmentation, stereo depth estimation, and temporal action lcoalization.
|
Updated: 2022-10-17