OpenPose has represented the first real-time multi-person system to jointly detect human body, hand, facial, and foot keypoints (in total 135 keypoints) on single images. It is authored by Ginés Hidalgo, Zhe Cao, Tomas Simon, Shih-En Wei, Yaadhav Raaj, Hanbyul Joo, and Yaser Sheikh.

Features

  • Whole-body 3D Pose Reconstruction and Estimation
  • Unity Plugin
  • Whole-body (Body, Foot, Face, and Hands) 2D Pose Estimation
  • Runtime Analysis
  • 2D real-time multi-person keypoint detection
  • 2x21-keypoint hand keypoint estimation
  • 3D triangulation from multiple single views
  • Unity Plugin
  • Runtime Analysis
  • Calibration toolbox: Estimation of distortion, intrinsic, and extrinsic camera parameters.
  • Synchronization of Flir cameras handled.
  • Single-person tracking
  • Python API
  • C++ API

Input

OpenPose supports image, video, webcam, Flir/Point Grey, IP camera, and support to add your own custom input source (e.g., depth camera).


Supported platforms

  1. Windows
  2. Linux (Ubuntu)
  3. macOS

Hardware compatibility

CUDA (Nvidia GPU), OpenCL (AMD GPU), and non-GPU (CPU-only) versions.


License

OpenPose is freely available for free non-commercial use, and may be redistributed under these conditions. Please, see the license for further details. Interested in a commercial license? Check this FlintBox link. For commercial queries, use the Contact section from the FlintBox link and also send a copy of that message to Yaser Sheikh.

Resources