LUCID: Learning Embodiment-Agnostic Intent Models from Unstructured Human Videos for Scalable Dexterous Robot Skill Acquisition figure
AlphaXiv 中文概览(可滚动查看)