facebookresearch/momentum
A library for human kinematic motion and numerical optimization solvers to apply human motion
What it solves
Momentum provides a set of foundational algorithms and numerical optimization solvers designed to handle human kinematic motion. It allows developers to apply human motion data and analysis to various applications, such as body tracking from different camera inputs.
How it works
The project implements human kinematic motion algorithms and solvers that can be used in both Python and C++. It offers differentiable solvers and geometry modules, with support for both CPU and GPU (via PyTorch) to handle the complex calculations required for motion tracking and kinematic parameterization.
Who it’s for
Researchers and developers working on human motion analysis, computer vision, and robotics who need robust tools for forward and inverse kinematics and body tracking.
Highlights
- Support for both Python and C++ APIs.
- Differentiable solvers for human kinematic motion.
- Capabilities for RGBD and monocular RGB body tracking.
- Flexible installation options via Conda, Pixi, and PyPI.
Related
- Project
- Project
- Project
- Project
- Project