EgoScale aligns human and robot dexterity data
NVIDIA's EgoScale work demonstrated an aligned human-robot data collection and transfer setup.
Why it matters
NVIDIA EgoScale used a Sharpa hand action space and MANUS/Vive capture for aligned human–robot data, with transfer tests on a Unitree G1 tri-finger hand.
Evidence notes
- The pipeline used a Sharpa hand action space with MANUS glove and Vive motion capture.
- Transfer tests used a Unitree G1 tri-finger hand configuration.
Company context
NVIDIA is a compute and AI infrastructure company providing the simulation, training, edge-compute and model platforms used across modern robotics. Its stack includes tools for robot learning, digital twins, perception and on-device inference.