# openpifpaf/openpifpaf

Official implementation of "OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association" in PyTorch.

Repository: https://github.com/openpifpaf/openpifpaf
Canonical: https://ross.abutalabs.com/products/openpifpaf
Homepage: https://openpifpaf.github.io
Language: Python
License: NOASSERTION
License Family: other
Topics: human-pose-estimation, keypoint-estimation, pose-estimation, computer-vision, composite-fields, deep-learning
Last push: 2024-08-15T18:43:47+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2750, "days_push": 748, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1261, forks 255 (observed 2026-08-28T04:04:10.093540+00:00)

## What it is
OpenPifPaf is a PyTorch library implementing Composite Fields for semantic keypoint detection and spatio-temporal association, primarily for human pose estimation and tracking. It provides real-time, single-stage detection and association of keypoints across images and video frames.

## Use cases
- detect human body pose in images
- track human pose across video frames in real time
- estimate keypoints for animals or vehicles
- run pose estimation on webcam or video streams
- benchmark pose estimation on COCO or PoseTrack datasets
- build perception for self-driving cars or delivery robots

## When to choose
- you need real-time pose detection and tracking in a single stage
- you want a well-tested Python/PyTorch pose estimation library
- you need keypoint detection beyond humans, such as animals or vehicles

## When to avoid
- you need general object detection or segmentation rather than keypoints
- you need a non-PyTorch framework or mobile-first deployment
- you require a permissively licensed library, as the license is non-standard

## Facets
- artifact type: library
- maturity: maintenance
- function: computer-vision, machine-learning, deep-learning, image-processing
- domain: computer-vision, machine-learning, deep-learning, artificial-intelligence
- platform: python, windows
- tags: pose-estimation, keypoint-detection, human-pose, pytorch, pose-tracking, real-time, linux, macos, gpu

## Member repositories
- openpifpaf/openpifpaf (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:10.093540+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:04:19.848446+00:00, confidence not recorded.
  - readme: https://github.com/openpifpaf/openpifpaf (fetched 2026-08-28T04:04:10.093540+00:00, sha 19f497780f24)
  - homepage: https://openpifpaf.github.io (fetched 2026-08-29T12:16:54.083368+00:00, sha 44136fa355b3)
- Data as of 2026-08-30T08:39:29.467469+00:00.