tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1371 stars · 405 forks observed · 2026-08-28

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

A PyTorch implementation of the CVPR'17 Realtime Multi-Person 2D Pose Estimation (OpenPose/rtpose) model. It provides pretrained weights, demo scripts for images and webcam, and training/evaluation code on the COCO keypoints dataset.

Use cases

• detect human body keypoints in images
• estimate poses of multiple people in real time from a webcam
• train a multi-person pose estimation model on COCO
• evaluate pose estimation mAP on COCO val2017
• port the original Caffe OpenPose model to PyTorch

When to choose

• you need multi-person 2D skeleton/keypoint detection in PyTorch
• you want a realtime pose estimation pipeline with pretrained rtpose weights
• you need reproducible training and evaluation on COCO keypoints

When to avoid

• you need 3D pose estimation or single-person-only models
• you need actively maintained code with a clear license
• you work without NVIDIA GPUs or on non-Linux platforms

Facets

library · maturity maintenance

machine-learning computer-vision image-processing deep-learning computer-vision machine-learning deep-learning python pose-estimation pytorch multi-person openpose coco keypoints human-pose linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="tensorboy/pytorch_Realtime_Multi-Person_Pose_Estimation")