CMU-Perceptual-Computing-Lab/openpose

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

34413 stars · 8037 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

OpenPose is a real-time multi-person keypoint detection library that jointly estimates human body, face, hand, and foot keypoints (135 total) from single images or video. It is written in C++ with Python bindings and built on Caffe and OpenCV.

Use cases

• detect human body keypoints in images or video in real time
• estimate facial landmarks and hand keypoints from photos
• track multiple people's poses in a video stream
• extract 2D pose data for action recognition or behavior analysis
• build motion capture or fitness form-checking applications
• annotate datasets with human pose keypoints

When to choose

• you need real-time multi-person 2D pose estimation including face, hands, and feet
• you want a well-established, widely cited pose estimation library with C++ and Python APIs
• you need to run pose estimation on CPU or NVIDIA GPUs across Linux, macOS, and Windows

When to avoid

• you need 3D pose estimation or state-of-the-art transformer-based accuracy
• you want a lightweight pure-Python or pip-installable solution
• your project requires a permissively licensed library, since OpenPose uses a non-commercial license
• you need actively developed features, as the project is largely in maintenance mode

Facets

library · maturity maintenance

computer-vision machine-learning deep-learning image-processing computer-vision machine-learning deep-learning image-processing windows cpp python pose-estimation keypoint-detection human-pose real-time multi-person face-landmarks hand-tracking caffe opencv linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="CMU-Perceptual-Computing-Lab/openpose")