|
Yinji ShenTu I am a Ph.D. student in the ZJU3DV group at Zhejiang University, where I have studied since 2024 under the supervision of Xiaowei Zhou and Wei Hua. I previously worked as a research assistant in the group. I received a B.S. from Dalian University of Technology and an M.Sc. from The University of Hong Kong. My research spans computer vision and graphics, with a focus on learning-based reconstruction and editing of volumetric video, including video reconstructed from monocular footage. Email / Google Scholar / GitHub / X |
|
ResearchSelected publications and academic service. Highlighted entries are representative publications. |
 |
DynOLAT: OLAT Video Prediction for Controllable Dynamic Full-Body Relighting
TL;DR: Predicts OLAT video for controllable relighting of dynamic full-body performances. My main contribution: leading the dynamic full-body OLAT dataset, covering 224 identity–outfit combinations. |
 |
2nd Volumetric Video Challenge
TL;DR: A workshop and challenge on volumetric video compression and sparse-view reconstruction. |
 |
RealMe: Immersive Volumetric Video from Minimal Captures
TL;DR: Recovers high-resolution volumetric video from sparse, low-resolution captures. |
 |
Track4D-Calib: Track-Driven Geometric Closure for 4D Camera-Array Calibration
TL;DR: Calibrates dynamic and static cameras in 4D capture systems with temporally coherent tracks. |
 |
Dyn-E: Local Appearance Editing of Dynamic Neural Radiance Fields
TL;DR: Edits the local appearance of a volumetric video from a single edited frame. |
Experience
4DV.ai — Algorithm Research and Development on Volumetric Video, Apr. 2025–Apr. 2026
Hikvision — Algorithm research intern; worked on CCTV image quality, May–Sep. 2022.
Awards
ChinaGraph 2024 Best Paper Award — Dyn-E: Local Appearance Editing of Dynamic Neural Radiance Fields.
|
Thanks Jon Barron for sharing this website's template source code. |