open-mmlab/mmpose

Health v2 · maintenance only

Adoption not part of the score

7855 stars · 1528 forks observed · 2026-08-28

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

MMPose is an open-source pose estimation toolbox and benchmark built on PyTorch as part of the OpenMMLab ecosystem. It provides implementations of state-of-the-art keypoint detection models for humans, animals, hands, and faces, along with training and evaluation pipelines.

Use cases

• estimate human pose keypoints from images
• train a pose estimation model on COCO or MPII
• benchmark keypoint detection models like HRNet and RTMPose
• detect hand keypoints for gesture recognition
• run face landmark detection
• estimate animal poses in videos
• fine-tune a pretrained pose model on a custom dataset

When to choose

• you need production-quality pose estimation with a large model zoo
• you want to train or benchmark keypoint detection models in PyTorch
• you need pose estimation for humans, animals, hands, or faces in one toolkit

When to avoid

• you need a lightweight inference-only solution without the OpenMMLab dependency stack
• your task is object detection or segmentation rather than keypoint estimation
• you work outside Python/PyTorch ecosystems

Facets

library · maturity active

machine-learning computer-vision image-processing benchmarking computer-vision deep-learning machine-learning image-processing python cross-platform pose-estimation keypoint-detection human-pose pytorch model-zoo rtmpose hand-pose-estimation face-keypoints animal-pose-estimation openmmlab gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="open-mmlab/mmpose")