VICAR: Video Imitation with Contact-Aware Retargeting for Humanoid Interactive Skills

Key points

• Third-person human videos offer a scalable alternative to task-specific reward engineering, teleoperation, and manually scripted demonstrations for humanoid loco-manipulation.1
• Recent video-to-humanoid systems have shown that noisy human motion, object trajectories, and contacts can be extracted from unstructured videos and refined into deployable skills.1
• Unlike retargeting methods that only remove penetration or preserve local interaction geometry, our formulation represents user-specified contact poses and collision clearance as high-priority Lagrangian penalty terms and optimizes the full motion to remain smooth through contact-rich transitions.1
• These results suggest that contact-preserving retargeting is a practical step toward humanoids that learn useful interactive behavior by watching humans.1

Sources (1)

• [1]VICAR: Video Imitation with Contact-Aware Retargeting for Humanoid Interactive Skills
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 06:34 AM

  Towards this, we present VICAR (Video Imitation with Contact-Aware Retargeting), a pipeline for learning humanoid interactive skills from third-person human videos.

  Third-person human videos offer a scalable alternative to task-specific reward engineering, teleoperation, and manually scripted demonstrations for humanoid loco-manipulation.

Extractive summary: sentences quoted from the sources.