LeCAR-Lab/human2humanoid

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1060 stars · 100 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Official implementation of H2O and OmniH2O, reinforcement-learning-based systems for real-time whole-body teleoperation of full-sized humanoid robots using human motion (VR, RGB camera, or verbal input). It includes training pipelines built on Isaac Gym, Legged Gym, and RSL RL, plus the OmniH2O-6 whole-body control dataset.

Use cases

• teleoperate a humanoid robot with a VR headset or RGB camera
• train RL policies for humanoid whole-body control in simulation
• learn autonomous humanoid skills from teleoperated demonstrations
• retarget human motion capture data to humanoid robots
• run sim-to-real deployment of humanoid locomotion policies

When to choose

• you need state-of-the-art human-to-humanoid whole-body teleoperation with a real-time control interface
• you want a research codebase for RL-based humanoid control on Isaac Gym
• you need a humanoid whole-body skill dataset for imitation learning

When to avoid

• you need a permissively licensed library for commercial products - the code is CC BY-NC 4.0
• you don't have access to NVIDIA Isaac Gym or a CUDA GPU
• you need a production-ready robotics stack rather than research code

Facets

library · maturity active

machine-learning reinforcement-learning simulation robotics robotics deep-learning autonomous-vehicles python humanoid-robotics teleoperation sim-to-real isaac-gym motion-capture research-code non-commercial-license gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="LeCAR-Lab/human2humanoid")