Dafei Qin | The University of Hong Kong

About Me

I am a Ph.D. student in the Department of Computer Science at The University of Hong Kong, supervised by Prof. Taku Komura. My research interest is Computer Graphics, with a specific focus on facial and body animation. I received my B.Sc. degree from the Department of Electronic Engineering, Tsinghua University.

News

[May. 2024] Start intern @ Adobe.
[May. 2024] One paper is accepted by SIGGRAPH 2024.
[Oct. 2024] Start intern @ Deemos Tech.
[May. 2023] Two papers are accepted by SIGGRAPH 2023.

Projects

Training Free Motion Intensity Control for Video Generation

Structural Match for Video Generation

Adobe MAX 2024

Li-Yi Wei

Publications

We translate PBR Facial assets to Gaussian Splatting counterpart in seconds, enabling 30fps@1400p rendering on mobile phones.

Instant Facial Gaussians Translator for Relightable and Interactable Facial Rendering

Dafei Qin, Hongyang Lin, Qixuan Zhang, Kaichun Qiao, Longwen Zhang, Zijun Zhao, Jun Saito, Jingyi Yu, Lan Xu, Taku Komura

arXiv

Project Page

We generate 3D talking face animation with head motions from audio, with rich controls from styles, text, images and etc.

Media2Face: Co-speech Facial Animation Generation With Multi-Modality Guidance

Qingcheng Zhao, Pengyu Long, Qixuan Zhang, Dafei Qin, Han Liang, Longwen Zhang, Yingliang Zhang, Jingyi Yu, Lan Xu

ACM SIGGRAPH 2024 (Conference Track)

Project Page

Top: Given an unrigged facial mesh with an unknown expression and identity in an arbitrary triangulation (yellow), NFR can transfer the expression to unrigged facial meshes with arbitrary triangulations (cyan). Bottom: NFR provides an interpretable latent space for user-friendly editing of the retargeted meshes.

Neural Face Rigging for Animating and Retargeting Facial Meshes in the Wild

Dafei Qin, Jun Saito, Noam Aigerman, Thibault Groueix, Taku Komura

ACM SIGGRAPH 2023 (Conference Track)

Project Page

Given an arbitrary input speech, our proposed transformer-based model, BodyFormer, can generate a sequence of vivid 3D body gestures

Bodyformer: Semantics-guided 3D Body Gesture Synthesis With Transformer

Kunkun Pang^*, Dafei Qin^*, Yingruo Fan, Julian Habekost, Takaaki Shiratori, Junichi Yamagishi, Taku Komura

ACM SIGGRAPH 2023 (Journal Track)
^*Joint First Authors

Awards and Scholarships

HKU Presidential PhD Scholarship, HKU, 2020-2024
University Scholarship, Tsinghua, 2019

Teaching Experience

Teaching assistant, COMP3271: Computer Graphics
Teaching assistant, ELEC2441: Computer Orgnization and Microprocessors