Embodied AI TopConf · CVPR2025

VidBot: Learning Generalizable 3D Actions from In-the-Wild 2D Human Videos for Zero-Shot Robotic Manipulation

CVPR2025 / 3D Vision

3D 视觉视频感知
新标签打开 AlphaXiv新标签打开 Project
VidBot: Learning Generalizable 3D Actions from In-the-Wild 2D Human Videos for Zero-Shot Robotic Manipulation figure
AlphaXiv 中文论文页面（可滚动查看）