NVIDIA's SOMA Retargeter adapts human motion to five humanoid robots, from G1 to AgiBot
jonstephens85 · x · 2026-11-20
NVIDIA released SOMA Retargeter, which adapts human motion data to the proportions and joint structures of specific humanoid robots via a shared SOMA skeleton.
- The pipeline scales motion targets, solves IK with Newton built on Warp, and applies corrections like foot-plant adjustment.
- Motion can come from mocap datasets (BONES-SEED), video reconstruction (GEM-X), or generative models outputting the SOMA skeleton.
- v0.2 ships reference configurations for Unitree G1/H2, Booster T1, AgiBot X2 Ultra and AgiBot A3T3; retargeted CSV trajectories feed policy-training workflows like SONIC and ProtoMotions.
More from Embodied
- GXO's 1,000-robot Apptronik deal is a letter of intent, hinging on 4 years of KPIs — YvesMulkers · 2026-11-20
- AFFECT from KAIST makes world models dynamics-faithful via action effect matching — John Won · 2026-11-20
- Mind2Act World: a 180-task benchmark exposing robots' mind-action composition failures — Mind2Act-World · 2026-11-20
- ZeroVLN navigates without SFT of the VLM brain using tool-calling and closed-loop RL — ZYT-AI · 2026-11-20
- Robot Paper Club recordings go live with four post-training talks; World Models edition next — DominiqueCAPaul · 2026-11-20
- Chinese student builds wearable tactical backpack robot, demo goes viral — TinfoilTricorn · 2026-11-20