xbpeng/DeepMimic

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3085 stars · 532 forks observed · 2026-08-28

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

DeepMimic is a C++ simulation framework with a Python (SWIG/TensorFlow) wrapper that trains simulated humanoid characters to imitate motion-capture skills using deep reinforcement learning. It implements the DeepMimic and AMP (Adversarial Motion Priors) research methods for physics-based character animation.

Use cases

• train a simulated humanoid to imitate mocap motion clips with reinforcement learning
• research physics-based character animation and motion imitation
• reproduce DeepMimic or AMP paper experiments
• build RL agents that perform locomotion, acrobatics, and martial arts skills in physics simulation
• study adversarial motion priors for stylized character control

When to choose

• you need the original reference implementation of the DeepMimic or AMP papers
• you are studying or extending physics-based motion imitation research in C++

When to avoid

• you want a maintained or easy-to-use implementation - the README says this codebase is deprecated in favor of MimicKit
• you need modern TensorFlow or a quick setup - it depends on old TensorFlow 1.x and manual C++ builds

Facets

library · maturity abandoned

reinforcement-learning simulation machine-learning graphics machine-learning simulation graphics robotics windows cpp python physics-based-animation motion-imitation mocap character-control deprecated bullet-physics tensorflow linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="xbpeng/DeepMimic")