Hi there! I’m currently a Researcher at Lenovo Research. I obtained my B.E. in Electronic Information Engineering from Xidian University in 2021, followed by a Master degree (Diplôme d’Ingénieur) in Multimedia Communications from Nantes Université, France in 2022. My research interests lie in Computer Vision, with a specific focus on Controllable Video/Image Generation and Editing. I previously have research experience in Video Compression (H265/VP9/AV1 Codec). I’m passionate about AIGC technology, feel free to contact me at xduzhangjiayu@163.com if you’d like to collaborate with me.
🔥 News
- 2025.11: 🎉🎉 One paper is accepted by TCSVT (CCF-B)!
- 2025.09: 🎉🎉 One paper about Controllable Video Generation is on Arxiv!
📝 Publications
DiTraj: Training-free Trajectory Control for Video Diffusion Transformer
(Co-first Author + Project Leader)
Cheng Lei†, Jiayu Zhang†‡, Yue Ma*, Xinyu Wang, Long Chen, Liang Tang, Yiqiang Yan, Fei Su, Zhicheng Zhao*
(†Equal Contribution ‡Project Lead *Corresponding Author)
- DiTraj is the first training-free trajectory control method for DiT-based video generation models!
📖 Experiences
- 2022.11 - now, Researcher, Lenovo Research
- 2020.09 - 2022.09, Master’s degree (Diplôme d’Ingénieur) in Multimedia Communications, Nantes Université, France
- 2017.09 - 2021.06, Bachelor of Engineering (B.E.) in Electronic Information Engineering, Xidian University, Xi’an, China
💻 Open-Source Contribution
- diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
- intel-npu-acceleration-library: A PyTorch acceleration library for the NPU of Intel Core Ultra processors.