Generated Reality: Human-centric World Simulation using Interactive Video Generation with Hand and Camera Control
Overall framework
Abstract
Generated Reality studies interactive human-centric world simulation for extended reality. The system conditions video generation on tracked head pose and joint-level hand poses, then distills a bidirectional video diffusion teacher into a causal interactive model for egocentric virtual environments.
Type
Publication
In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Findings
Tong WU 吴桐
Assistant Professor @ Fudan
My research interests include 3d vision, long-tailed recognition, and robustness.