HDMI open-sourced: whole-body robot interaction skills learned from human videos
chris_j_paxton · x · 2026-09-08
Elijah Galahad open-sourced HDMI, a general framework for learning whole-body robot interaction skills directly from human videos — no manual reward engineering or task-specific pipelines required.
- Demonstrated 67 door traversals in the real world across 6 real tasks, plus 14 simulated tasks
- HDMI extends MimicLite from robot-only motion tracking to paired robot-object interaction tracking
- The rewritten release adds a new Mjlab backend and single-object multi-motion support
- The GitHub README includes full uv + MjLab setup and reproduction commands, with W&B and HuggingFace credentials configuration
More from Embodied
- RealSense D585 Pro AI-Powered 3D Stereo Depth Camera Unboxed for Physical AI — chrismatthieu · 2026-10-26
- Physical world agent swarm demo shared on X — pduan · 2026-10-26
- Looking for a wearable mic with best battery life that feeds recordings to personal agents via MCP — altryne · 2026-10-26
- 'Lamp Swarm': a desk of lamp-shaped robots shows what physical AI work looks like — dee_hw · 2026-10-26
- Tencent's HiSTEVE 1.0: hierarchical VLA agent plays Minecraft with native keyboard and mouse, hits 85% SOTA — _akhaliq · 2026-10-26
- SJTU's VaM lets robots learn online from mistakes, 95%+ success over 4-hour autonomous runs — chris_j_paxton · 2026-10-26