# Daniil-Osokin/lightweight-human-pose-estimation.pytorch

Fast and accurate human pose estimation in PyTorch. Contains implementation of "Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose" paper.

Repository: https://github.com/Daniil-Osokin/lightweight-human-pose-estimation.pytorch
Canonical: https://ross.abutalabs.com/products/lightweight-human-pose-estimationpytorch
Language: Python
License: Apache-2.0
License Family: permissive
Topics: human-pose-estimation, deep-learning, real-time, openpose, openvino, coco-keypoints-detection, mscoco-keypoint, lightweight, pytorch, lightweight-openpose, pose-estimation
Last push: 2024-04-30T09:54:40+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2741, "days_push": 855, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2241, forks 495 (observed 2026-08-28T04:06:29.835851+00:00)

## What it is
A PyTorch implementation of Lightweight OpenPose for real-time 2D multi-person human pose estimation on CPU. It detects up to 18 body keypoints per person in an image and includes training, validation, and demo code.

## Use cases
- detect human skeletons in images
- run real-time pose estimation on CPU
- train a lightweight OpenPose model on COCO
- estimate body keypoints for multiple people
- export pose estimation models for OpenVINO
- build fitness or motion analysis apps

## When to choose
- you need fast multi-person 2D pose estimation without a GPU
- you want a reproducible PyTorch training pipeline for Lightweight OpenPose
- you target CPU deployment with OpenVINO

## When to avoid
- you need 3D pose estimation
- you require state-of-the-art accuracy on COCO keypoints
- you need a maintained library with recent framework support

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning, image-processing
- platform: python
- tags: pose-estimation, openpose, keypoint-detection, pytorch, real-time-inference, cpu-inference, openvino, coco-dataset, linux, gpu

## Member repositories
- Daniil-Osokin/lightweight-human-pose-estimation.pytorch (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:29.835851+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 2026-08-30T02:44:05.765464+00:00, confidence not recorded.
  - readme: https://github.com/Daniil-Osokin/lightweight-human-pose-estimation.pytorch (fetched 2026-08-28T04:06:29.835851+00:00, sha 578dde7ee1e5)
- Data as of 2026-08-30T08:39:29.467469+00:00.