Do As I Do retargets internet and generated videos into executable robot hand data
chris_j_paxton · x · 2026-09-25
Scaling dexterous manipulation data is a key robotics bottleneck: real-world finger-pose data is often subtly off, physically invalid, and hard to execute or learn from. RoboPapers Episode 107 covers Do As I Do, an algorithm that reconstructs and retargets monocular RGB videos — including internet and AI-generated clips — into high-quality, executable robot data for high-dof hands, outperforming the state of the art.
Related event: Berkeley's Do as I Do Trains Dexterous Hands from Human Video(3 posts)→
More from Embodied
- Robot fails at folding a towel, reassuring roboticists frontier models won't eat robotics yet — chris_j_paxton · 2026-11-12
- Nuro podcast: manufacturing, charging, and regulation gate robotaxi mass adoption — R_D · 2026-11-12
- Cybercab rear-ended by human driver, researcher quips: the problem is the humans — chris_j_paxton · 2026-11-12
- HT-Policy Uses Heavy-Tailed Student-t Regression to Close Robot Policy Gap — chris_j_paxton · 2026-11-12
- NVIDIA GTC Berlin Side Meetup Balloons to 2,000 Signups for a 500-Cap Brewery — lukas_m_ziegler · 2026-11-12
- Lamport Robotics' Pincer1 Robotic Arm Priced at $1,199, Payable Only in SOL or USDC — R4nsack · 2026-11-12