Sotabase

Career

· Research Intern, ByteDance Seed2025–
· Research Intern, NVIDIA2025–
· Research Intern, Adobe Research2023–
· PhD in Computer Science, Stanford University2023–
· Visiting Student Researcher, Stanford University2022–
· Research Student, ETH Zürich CVL & Toyota TRACE2021–
· MS in Computer Science, ETH Zürich2020–2023
· Bachelor of Science (Honours) in Computer Science, King's College London2017–2020

Publications (34)

Pix2NeRF: Unsupervised Conditional $\pi$-GAN for Single Image to Neural Radiance Fields Translation
Computer Vision and Pattern Recognition · 2022
109
cited
Collaborative Video Diffusion: Consistent Multi-video Generation with Camera Control
Neural Information Processing Systems · 2024
89
cited
Frame Context Packing and Drift Prevention in Next-Frame-Prediction Video Diffusion Models
2025
59
cited
DiffDreamer: Towards Consistent Unsupervised Single-view Scene Extrapolation with Conditional Diffusion Models
IEEE International Conference on Computer Vision · 2022
53
cited
Diffusion Self-Distillation for Zero-Shot Customized Image Generation
Computer Vision and Pattern Recognition · 2024
39
cited
Mixture of Contexts for Long Video Generation
arXiv.org · 2025
32
cited
Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models
Computer Vision and Pattern Recognition · 2023
26
cited
Captain Cinema: Towards Short Movie Generation
arXiv.org · 2025
20
cited
X-Dyna: Expressive Dynamic Human Image Animation
Computer Vision and Pattern Recognition · 2025
14
cited
DiffDreamer: Consistent Single-view Perpetual View Generation with Conditional Diffusion Models
arXiv.org · 2022
13
cited
Robust Symmetry Detection via Riemannian Langevin Dynamics
ACM SIGGRAPH Conference and Exhibition on Computer Graphics and Interactive Techniques in Asia · 2024
13
cited
ByteMorph: Benchmarking Instruction-Guided Image Editing with Non-Rigid Motions
arXiv.org · 2025
6
cited
CL-Splats: Continual Learning of Gaussian Splatting with Local Optimization
arXiv.org · 2025
6
cited
Pretraining Frame Preservation in Autoregressive Video Memory Compression
arXiv.org · 2025
3
cited
BulletTime: Decoupled Control of Time and Camera Pose for Video Generation
arXiv.org · 2025
1
cited
Scene-Level Appearance Transfer with Semantic Correspondences
International Conference on Computer Graphics and Interactive Techniques · 2025
1
cited
ReStyle3D: Scene-Level Appearance Transfer with Semantic Correspondences
arXiv.org · 2025
Sotabase
Shengqu Cai | Researcher Profile | Sotabase | Sotabase