📚 Publications

RMD: A Simple Baseline for More General Human Motion Generation via Training-free Retrieval-Augmented Motion Diffuse

Zhouyingcheng Liao, Mingyuan Zhang, Wenjia Wang, Lei Yang, Taku Komura

Arxiv 2024

[paper]

A training-free retrieval-augmented system to generate 3D motion from text.

It Takes Two: Real-time Co-Speech Two-person's Interaction Generation via Reactive Auto-regressive Diffusion Model

Mingyi Shi, Dafei Qin, Leo Ho, Zhouyingcheng Liao, Yinghao Huang, Junichi Yamagishi, Taku Komura

Arxiv 2024

[paper]

An online system for generating dynamic, interactive motions for two characters from speech audio.

SIMS: Simulating Human-Scene Interactions with Real World Script Planning

Wenjia Wang, Liang Pan, Zhiyang Dou, Jidong Mei, Zhouyingcheng Liao, Yuke Lou, Yifan Wu, Lei Yang, Jingbo Wang, Taku Komura

Arxiv 2024

[paper]

A novel framework combining script-driven intent via Retrieval-Augmented Generation with a versatile physics-based control policy to generate human-scene interactions.

SENC: Handling Self-collision in Neural Cloth Simulation

Zhouyingcheng Liao*, Sinan Wang*, Taku Komura

ECCV 2024

[Webpage] [code]

A self-supervised neural cloth simulator that effectively addresses cloth self-collision.

EMDM: Efficient Motion Diffusion Model for Fast, High-Quality Human Motion Generation

Wenyang Zhou, Zhiyang Dou, Zeyu Cao, Zhouyingcheng Liao, Jingbo Wang, Wenjia Wang, Yuan Liu, Taku Komura, Wenping Wang, Lingjie Liu

ECCV 2024

[webpage] [paper] [video]

A fast and high-quality human motion generation method, which takes only 0.05s for a sequence of 196 frames.

VINECS: Video-based Neural Character Skinning

Zhouyingcheng Liao, Vladislav Golyanik, Marc Habermann, Christian Theobalt

CVPR 2024

[paper]

The first end-to-end method for generating a dense and rigged 3D character mesh with learned pose-dependent skinning weights solely from multi-view videos.

Skeleton-free Pose Transfer for Stylized 3D Characters

Zhouyingcheng Liao, Jimei Yang, Jun Saito, Gerard Pons-Moll, Yang Zhou

ECCV 2022

[webpage] [paper] [code]

The first neural method that achieves automatic pose transfer between any stylized 3D characters, without any rigging, skinning or manual correspondence.

TailorNet: Predicting Clothing in 3d as a Function of Human Pose, Shape and Garment Style

Chaitanya Patel*, Zhouyingcheng Liao*, Gerard Pons-Moll (*: co-first author)

CVPR 2020 oral

[webpage] [paper] [video] [code] [data]

TailorNet is the first neural model which predicts clothing deformation in 3D as a function of three factors: pose, shape and style (garment geometry), while retaining wrinkle detail. We also present a garment animation dataset of 55800 frames generated by physically based simulation

Live Face Verification with Multiple Instantialized Local Homographic Parameterization.

Chen Lin, Zhouyingcheng Liao, Peng Zhou, Jianguo Hu, Bingbing Ni

IJCAI 2018

[paper]

Uniface: A Unified Network for Face Detection and Recognition

Zhouyingcheng Liao, Peng Zhou, Qinlong Wu, Bingbing Ni

ICPR 2018

[paper]

📚 Publications

🎓 Academic Service

Reviewer

Teaching Assistant