BIOGRAPHY
Yifan Zhou is an Ph.D. student at the School of Computer Science and Engineering, Nanyang Technological University, Singapore. He is a member of NTU MMLab affiliated with the NTU S-Lab, supervised by Professor Xingang Pan. Before that, he worked as research engineer in NTU S-Lab and Shanghai AI Lab. He received his bachelor degree at Computer Science School, Beijing Institute of Technology. His current research interest is image and video generation.
Switch to my Chinese blog to get more information.
EDUCATON
Aug 2023 - present, S-Lab, School of Computer Science and Engineering
Nanyang Technological University, Singapore
Sept 2017 – Jun 2021, Computer Science & Technology, School of Computer Science & Technology
Beijing Institute of Technology, China, GPA: 3.7/4.0 (88/100)
PROFESSIONAL EXPERIENCE
Research Engineer @ NTU S-Lab (Mar 2022 - Aug 2023)
- Participate in the development of an AI face editing project. I was responsible for model deployment and codebase management.
- Participate in a diffusion-based video translation project. I was mainly responsible for code implementation.
Algorithm Engineer @ Shanghai AI Lab (Jul 2021 - Feb 2022)
- Core developer and maintainer of MMDeploy, an open source deep learning model deployment library.
PUBLICATIONS
Trainable Log-linear Sparse Attention for Efficient Diffusion Transformers [PDF]
Yifan Zhou, Zeqi Xiao, Tianyi Wei, Shuai Yang, and Xingang Pan
- arXiv 2025
Alias-free latent diffusion models: Improving fractional shift equivariance of diffusion latent space [PDF]
Yifan Zhou, Zeqi Xiao, Shuai Yang, and Xingang Pan
- IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR, Oral), 2025
FRESCO: Spatial-Temporal Correspondence for Zero-Shot Video Translation [PDF]
Shuai Yang, Yifan Zhou, Ziwei Liu, Chen Change Loy
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
DiffMorpher: Unleashing the Capability of Diffusion Models for Image Morphing [PDF]
Kaiwen Zhang, Yifan Zhou, Xudong Xu, Xingang Pan, and Bo Dai
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
Rerender a video: Zero-shot text-guided video-to-video translation [PDF]
Shuai Yang, Yifan Zhou, Ziwei Liu, Chen Change Loy
ACM SIGGRAPH Asia 2023 Conference Proceedings
PanoFlow: Learning 360◦ Optical Flow for Surrounding Temporal Understanding [PDF]
Hao Shi†, Yifan Zhou†, Kailun Yang, Xiaoting Yin, Ze Wang, Yaozu Ye, Zhe Yin, Shi Meng, Peng Li, and Kaiwei Wang
IEEE Transactions on Intelligent Transportation Systems (ITITS)
Learning optical flow via cross strip correlation for autonomous driving [PDF]
Hao Shi,Yifan Zhou, Kailun Yang, Xiaoting Yin, and Kaiwei Wang
IEEE Intelligent Vehicles Symposium (IV)
HONORS/AWARDS
Programming Contests
ACM-ICPC: 1 Gold Medal, 5 Sliver Medal
CCPC(China Collegiate Programming Contest): 2 Sliver Medal
PROJECTS
Undergraduate Capstone: Efficient Gradient-domain Image Seamless Blending (Mar 2021 - Jun 2021)

blog link: https://zhouyifan.net/2021/05/10/20210322-PoissonImgaeEditImplementation/
github link: https://github.com/SingleZombie/Gradient-Domain-Image-Processing-Cpp
Image Segmentation using Superpixels(Jan 2020)
![]()
![]()
blog link: https://zhouyifan.net/2020/02/01/DIP-image-segmentation-project-2/
github link: https://github.com/SingleZombie/SLICAP-image-segmentation-cpp
Effective Sudoku Solver (Jan 2020)
blog link: https://zhouyifan.net/2019/12/23/Software-Engineering-Project/
github link: https://github.com/SingleZombie/sudoku-solver
Chinese Spelling Checker(Dec 2019)
blog link: https://zhouyifan.net/2020/02/09/Chinese-Spelling-Checker/
github link: https://github.com/SingleZombie/Simple-Chinese-Spelling-Checker
Ray tracing running on CPU(Aug 2019)

blog link: https://zhouyifan.net/2020/02/08/Cpp-Ray-Tracing-running-on-CPU/
github link: https://github.com/SingleZombie/Ray-Tracing-OpenGL
SKILLS/OTHERS
Computer Skills
- Rich experience in C++ programming(since 2014). Familiar with C++ STL, OpenGL, OpenCV, cocos2dx.
- Have accomplished projects using java, python.
- Learn and implement efficient algorithm quickly and correctly.
- Able to create animation by Flash and modify video by PR.