Xiaodong Cun – Researcher in Computer Photography

Xiaodong Cun（寸晓东）

Assistant Professor

Great Bay University, Dongguan, China

Email (cun AT gbu.edu.cn) / Google Scholar / Github / ZhiHu / CV / Official Homepage / GVC Lab

I am currently a tenure-track assistant professor in the SCIT(Computer Science) , Great Bay University since 2024.09. I received my M.S and Ph.D. degrees from the Department of Computer and Information Science, FST, University of Macau, under the supervision of Prof. Chi-Man Pun in 2018 and 2021 respectively. I received my B.Sc degree in Computer Science from Xidian University. My main research interests include Visual Computing, Computational Photography, with the intersection of Computer Vision and Graphics.

I am building Geneartive & intelligent Visual Computing Lab (GVC Lab), a new research lab at Great Bay University. Our lab studies Visual Computing, with the goal of building intelligent machines that for creators and ordinary users.
Several positions are available 🔥 🔥, please drop me an Email if you are interested.

News

3 papers are conditional accepted by SIGGRAPH Asia 2025.
2025.08
1 paper is accepted by ACM MultiMedia 2025.
2025.07
1 paper is conditional accepted by SIGGRAPH 2025.
2025.03.29
3 papers are accepted by CVPR 2025.
2025.02
MagicStick is accepted by WACV 2025 and CustomTTT is accepted by AAAI 2025.
2024.12
We won the 1st place in OOD-CV: SSB Challenge 2024 (Open-Set Recognition Track) at ECCV 24 Workshop.
2024.09
I officially leave Tencent (24/09/20) and join GBU (24/09/23) 🎉🎉.
2024.09
CV-VAE is accepted by NeurIPS 2024.
2024.09
3 papers are accepted by ECCV 2024 and 1 paper is accepted by Pattern Recognition.
2024.07
6 papers are accepted by CVPR 2024. (🌟 SmartEdit has been selected as highlight paper.)
2024.02
3 papers are accepted by AAAI 2024, ICLR 2024 (Spotlight) and Eurographics 2024, respectively.
2023.12
1 paper is accepted by NeurIPS 2023.
2023.09
1 paper is conditionally accepted by SIGGRAPH Asia 2023 (Conference Track).
2023.08
4 papers (4/5) are accepted by ICCV-23. (🌟 FateZero has been selected as Oral paper.)
2023.07
6 papers (6/7) are accepted by CVPR-23.
2023.02

Education and Work Experience

Assistant Professor, School of Computing and Information Technology, Great Bay University
2024.9 -
Senior Researcher, Visual Computing Center, Tencent AI Lab, lead by Dr. Jue Wang, Dr. Ying Shan.
2021.9 - 2024.9
Researcher (Intern), Huawei Noah's Ark Lab, work with Prof. Jianzhuang Liu.
2020.3-11
Asian University Machine Learning Camp, Jeju National University.
2018.9
Algorithm Engineer (Intern), HIKVision Research Institute.
2018.5-7
FrondEnd Engineer (Intern), Alibaba Group.
2014.7-9
Master & Ph.D, University of Macau, Computer Science. Supervisor: Prof. Chi-Man Pun.
2015.8 - 2021.9
Bachelor, Xidian University, Computer Science.
2011.8 - 2015.7

Research Highlight [Full Publications] (🧑‍💻 Co-first Author, 📮 Corresponding Author)

SadTalker: Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animation

Wenxuan Zhang 🧑‍💻, Xiaodong Cun 🧑‍💻, Xuan Wang, Yong Zhang, Xi Shen, Yu Guo, Ying Shan, Fei Wang

Computer Vision and Pattern Recognition (CVPR), 2023

arXiv / Online Demo / Project Page / Github

Top 10 you won't miss paper of CVPR 2023 (voxel51.com).
Top 10 Most Github Star CVPR paper (github.com).

VideoReTalking: Audio-based Lip Synchronization for Talking Head Video Editing In the Wild

Kun Cheng 🧑‍💻, Xiaodong Cun 🧑‍💻📮, Yong Zhang, Menghan Xia, Fei Yin,
Mingrui Zhu, Xuan Wang, Jue Wang, Nannan Wang

SIGGRAPH Asia (Conference Track), 2022

arXiv / Project Page / Github

Top 10 Most Github Star SIGGRAPH paper (github.com).

VideoCrafter1 & VideoCrafter2

Haoxin Chen 🧑‍💻, Menghan Xia 🧑‍💻, Yong Zhang 🧑‍💻, Xiaodong Cun🧑‍💻,Xintao Wang, Ying Shan

Computer Vision and Pattern Recognition (CVPR) & Technical report, 2024

arXiv / Project Page / Github

PaperDigest Most Influential Papers of ArXiv 24 (paperdigest.org).

FateZero: Fusing Attentions for Zero-shot Text-based Video Editing

Chenyang Qi, Xiaodong Cun 📮, Yong Zhang, Chenyang Lei, Xintao Wang, Ying Shan, Qifeng Chen 📮

International Conference on Computer Vision (ICCV), 2023

arXiv / Project Page / Github

ICCV 23 Oral Presentation (2.3%)
PaperDigest Most Influential Papers of ICCV 23 (paperdigest.org).

Uformer: A General U-Shaped Transformer for Image Restoration

Zhendong Wang, Xiaodong Cun 📮, Jianmin Bao, Jianzhuang Liu, Wengang Zhou, Houqiang Li

Computer Vision and Pattern Recognition (CVPR), 2022

arXiv / Github

PaperDigest Most Influential Papers of CVPR 22 (paperdigest.org).

Follow Your Pose: Pose-Guided Text-to-Video Generation using Pose-Free Videos

Yue Ma 🧑‍💻, Yingqing He 🧑‍💻, Xiaodong Cun, Xintao Wang, Ying Shan, Xiu Li , Qifeng Chen

AAAI Conference on Artificial Intelligence (AAAI ), 2024

arXiv / Project Page / Github

PaperDigest Most Influential Papers of AAAI 24 (paperdigest.org).

Talks

2025.08.21, Acceleration of Video Diffusion Model,VIVO VCAN Talk.
2025.08.16, The Magic of Video Diffusion Model: from cinemagraph to Short Story Generation, University of Macau.
2025.08.07, Towards Cinematic Video Generation, Tencent ARC Lab.
2025.07, GBU AI Talk.
2025.05.30, Video Animation and Relighting, VIVO Blue Talk.
2023.9.30, Applications of AIGC, Kwai (Kuaishou) Technology. Hosted by Dr. Xin Tao.
2023.6.30, AIGC related Applications, HIT SZ. Hosted by Prof. Xuebo Liu.
2023.6.16, Unlocking the ability of large text-to-image model: new applications and new opportunities, University of Macau. Hosted by Prof. Chi-Man Pun.

Misc

🎓 If you are looking for talent student professor on Machine Translation, please checkout my friend Xuebo Liu.
🎴 Level 19 (failed to level 20) in Slay the Spire. More than 10k death in Celeste.
🏃 Workout often & lose weight 10kg 7.5kg & 200+ hours in Fitness Boxing
🎮 Cloud player of DOTA 2, ancient player of Freestyle, just ordinary player.
👨‍💻 Out of fashion front-end engineer. Try this website template at Github.
💪 I also (try to) practice Slow Science, inspired by Alexei (Alyosha) Efros.
🌈 The color of this page is Hermes Orange.
💵 Lose the wallet of 0.2 BTC in 2013.
☕️ A big fan of Coffee and Tea.
❤️ Since 2019.4.13