Scholar
Shengqu Cai
Google Scholar ID: KeI51t8AAAAJ
Stanford University
Citations & Impact
Citations
298
H-index
6
i10-index
6
Publications
12
Co-authors
10
list available
Publications
Resume (English only)
Academic Achievements
- 1. Mixture of Contexts for Long Video Generation, arXiv, 2025.
- 2. FramePack: Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Models, NeurIPS 2025 (Spotlight).
- 3. Diffusion Self-Distillation for Zero-Shot Customized Image Generation, CVPR 2025.
- 4. Captain Cinema: Towards Short Movie Generation, arXiv, 2025.
- 5. CL-Splats: Continual Learning of Gaussian Splatting with Local Optimization, ICCV 2025.
- 6. Pix2NeRF, CVPR 2022.
Research Experience
- 1. Since September 2023, pursuing a PhD in Computer Science at Stanford University.
- 2. Summer 2023, research intern at Adobe.
- 3. 2022-2023, worked on diffusion with Eric Chan and Songyou Peng.
- 4. Started research career in 2021 working on NeRFs and GANs, mentored by Anton Obukhov.
Education
- 1. Stanford University, PhD in Computer Science, advised by Prof. Gordon Wetzstein and Prof. Leonidas Guibas, affiliated with Computational Imaging Lab and Geometric Computing Lab.
- 2. ETH Zürich, Master in Computer Science, supervised by Prof. Luc Van Gool.
- 3. King's College London, Bachelor in Computer Science with first honour, worked on information theory.
Background
- Research Interests: Solving tasks that are fundamentally ill-posed via traditional methods, focusing on generative models, primarily on video generation, hoping to one day forward simulate the future and reverse engineer the past. Enjoys making cool theories, videos, demos, and applications.
Miscellany
- Personal Interests: Enjoys making cool theories, videos, demos, and applications.
