About Me

I am currently a fourth-year Ph.D. student at Multimedia Computing Group (MCG) in Department of Computer Science and Technology, Nanjing University, supervised by Prof. Limin Wang. Before that, I received my bachelor degree from Nanjing University.

My current research interests lie in multimodal understanding and generation, including image/video grounding and synthesis.

🔥 News

• 2025.06: 🎉🎉 One first-authored paper (IBQ) accepted by ICCV 2025.
• 2024.11: We release SEED-Voken project: a Series of Powerful Visual Tokenizers.
• 2024.06: We release the code of Open-MAGVIT-2, which is an open source implementation of Google’s MAGVIT-2.
• 2024.02: 🎉🎉 One first-authored paper (PRVG) accepted by CVIU.
• 2024.02: 🎉🎉 One first-authored paper (BIVDiff) accepted by CVPR 2024.
• 2023.10: 🎉🎉 One first-authored paper (Dynamic MDETR) accepted by T-PAMI.

📝 Selected Publications

Check out full publication list at my Google Scholar profile.

ICCV 2025

Scalable Image Tokenization with Index Backpropagation Quantization
International Conference on Computer Vision, 2025
Fengyuan Shi*, Zhuoyan Luo*, Yixiao Ge, Yujiu Yang, Ying Shan, Limin Wang
[ArXiv][Code]

CVPR 2024

BIVDiff: A Training-Free Framework for General-Purpose Video Synthesis via Bridging Image and Video Diffusion Models
The IEEE Conference on Computer Vision and Pattern Recognition, 2024
Fengyuan Shi, Jiaxi Gu, Hang Xu, Songcen Xu, Wei Zhang, Limin Wang
[ArXiv][Project Page][Code]

T-PAMI 2024

Dynamic MDETR: A Dynamic Multimodal Transformer Decoder for Visual Grounding
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024
Fengyuan Shi, Ruopeng Gao, Weilin Huang, Limin Wang
[ArXiv][Code]

CVIU 2024

End-to-end dense video grounding via parallel regression
Computer Vision and Image Understanding, 2024
Fengyuan Shi, Weilin Huang, Limin Wang
[ArXiv][Code]

📝 Projects

Github Project

SEED-Voken: A Series of Powerful Visual Tokenizers
The project aims to provide advanced visual tokenizers for autoregressive visual generation and currently supports the following methods:

[Code]

• Open-MAGVIT2: An Open-source Project Toward Democratizing Auto-Regressive Visual Generation
  Zhuoyan Luo*, Fengyuan Shi*, Yixiao Ge, Yujiu Yang, Limin Wang, Ying Shan
  
• IBQ: Scalable Image Tokenization with Index Backpropagation Quantization
  Fengyuan Shi*, Zhuoyan Luo*, Yixiao Ge, Yujiu Yang, Ying Shan, Limin Wang
  

🎖 Honors and Awards

• Outstanding Graduate Student of Najing University, 2024
• National Scholarship, 2024
• President’s Special Scholarship for Doctoral Students of Nanjing University, 2021
• YeePay Scholarship, 2018
• Renmin Scholarship, 2017, 2019

📖 Educations

• 2021.09 - Now, Ph.D. in Computer Science and Technology, Nanjing University.
• 2017.09 - 2021.06, B.Sc. in Computer Science and Technology, Nanjing University.

🪪 Academic Service

• Conference Reviewer: CVPR 2024/2025, ICCV 2025, MM 2025, NeurIPS 2025

• Journal Reviewer: Pattern Recognition, Neural Networks, IEEE Transactions On Circuits and Systems, IET Computer Vision, IEEE Transactions on Circuits and Systems for Video Technology

💻 Internships

• 2025.02 - 2025.05, Research Intern, ByteDance Doubao Seed Team, Beijing, advised by Zhi Tian and Qiushan Guo.
• 2024.03 - 2025.01, Research Intern, Tencent ARC Lab, Beijing, advised by Yixiao Ge.
• 2023.07 - 2024.02, Research Intern, Huawei Noah’s Ark Lab, Shanghai, advised by Hang Xu.
• 2023.04 - 2023.06, Research Intern, Ant Group, Hangzhou.