PRISM scales humanoid loco-manipulation with counterfactual video generation, no real-world fine-tuning
Zihan Wang · hf · 2026-10-01
PRISM is a real-to-sim-to-real framework that overcomes the scarcity of high-quality human-object interaction videos for teaching humanoids loco-manipulation via visual imitation.
Pipeline:
- Generates hundreds of diverse "counterfactual" human-object interaction videos from a few real exemplars via video-to-video generation;
- A contact-anchored real-to-sim pipeline reconstructs human and object motions, retargeting imperfect video data into physically plausible trajectories;
- Intra-class variability across counterfactual videos enables a single policy that generalizes to unseen objects.
Using only onboard depth observations, the humanoid picks up, carries, and drops boxes, barrels, bins, and balls across novel instances and configurations, with zero real-world fine-tuning.
Related event: PRISM Trains Humanoid Robots with Counterfactual Video Data(2 posts)→
More from Embodied
- YC-backed Ornadyne's flapping-wing drone was classified as a bird by Air Force detectors — import_jmr · 2026-11-16
- Uber starts offering Amazon's Zoox robotaxi rides in Las Vegas — Polymarket · 2026-11-16
- Ant Lingbo Embodied AI Challenge crowns TARS team after 48-hour real-robot hackathon — 新智元 · 2026-11-16
- Tesla FSD fleet tops 7 billion city miles, adding ~500M in under four weeks — XFreeze · 2026-11-16
- Ropedia Releases Xperience 2: 1,500+ Scanned Real-World Scenes Paired With Egocentric Demos — liuziwei7 · 2026-11-16
- Hangzhou's Manycore Tech launches Aholo MS07 3D laser scanner with CHCNAV — teortaxesTex · 2026-11-16