ZheC/Realtime_Multi-Person_Pose_Estimation

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

5123 stars · 1351 forks observed · 2026-08-28

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

Reference implementation of the CVPR'17 paper 'Realtime Multi-Person Pose Estimation', a bottom-up approach that detects keypoints for multiple people in real time without a person detector. Built on Caffe with Python/Matlab/C++ tooling, it is the original codebase behind OpenPose.

Use cases

• detect human body keypoints in images
• estimate poses of multiple people in realtime video
• reproduce the CVPR 2017 OpenPose paper results
• train a multi-person pose estimation model on COCO keypoints
• run pose estimation demos on webcam footage

When to choose

• you need the original reference implementation of the CVPR'17 bottom-up pose estimation method
• you are doing research on multi-person pose estimation and want the paper's exact code
• you want to study or port the Caffe-based OpenPose predecessor

When to avoid

• you need a production-ready or actively maintained pose estimation library
• you prefer modern frameworks like PyTorch or TensorFlow - use the community reimplementations or OpenPose instead
• you need permissive licensing - the license is non-standard

Facets

library · maturity maintenance

computer-vision deep-learning image-processing computer-vision deep-learning machine-learning python cpp windows human-pose-estimation openpose caffe realtime cvpr-2017 keypoint-detection research-code linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ZheC/Realtime_Multi-Person_Pose_Estimation")