Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/openpifpaf/openpifpaf
Official implementation of "OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association" in PyTorch.
https://github.com/openpifpaf/openpifpaf
composite-fields computer-vision deep-learning human-pose-estimation keypoint-estimation pose-estimation
Last synced: 29 days ago
JSON representation
Official implementation of "OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association" in PyTorch.
- Host: GitHub
- URL: https://github.com/openpifpaf/openpifpaf
- Owner: openpifpaf
- License: other
- Created: 2019-02-21T22:07:00.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2024-08-15T18:43:47.000Z (3 months ago)
- Last Synced: 2024-10-01T15:02:55.200Z (about 1 month ago)
- Topics: composite-fields, computer-vision, deep-learning, human-pose-estimation, keypoint-estimation, pose-estimation
- Language: Python
- Homepage: https://openpifpaf.github.io
- Size: 53.5 MB
- Stars: 1,168
- Watchers: 33
- Forks: 246
- Open Issues: 49
-
Metadata Files:
- Readme: README.md
- Contributing: docs/CONTRIBUTING.md
- License: LICENSE
- Citation: citation.bib
Awesome Lists containing this project
- awesome-human-pose-estimation - [code - CVPR 19, PifPaf (Multi-Person 2D Pose Estimation / 2019)
README
# openpifpaf
Continuously tested on Linux, MacOS and Windows:
[](https://github.com/openpifpaf/openpifpaf/actions/workflows/tests.yml)
[](https://github.com/openpifpaf/openpifpaf/actions/workflows/deploy-guide-dev.yml)
[](https://pepy.tech/project/openpifpaf)
[__New__ 2021 paper](https://arxiv.org/abs/2103.02440):
> __OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association__
> _[Sven Kreiss](https://www.svenkreiss.com), [Lorenzo Bertoni](https://scholar.google.com/citations?user=f-4YHeMAAAAJ&hl=en), [Alexandre Alahi](https://scholar.google.com/citations?user=UIhXQ64AAAAJ&hl=en)_, 2021.
>
> Many image-based perception tasks can be formulated as detecting, associating
> and tracking semantic keypoints, e.g., human body pose estimation and tracking.
> In this work, we present a general framework that jointly detects and forms
> spatio-temporal keypoint associations in a single stage, making this the first
> real-time pose detection and tracking algorithm. We present a generic neural
> network architecture that uses Composite Fields to detect and construct a
> spatio-temporal pose which is a single, connected graph whose nodes are the
> semantic keypoints (e.g., a person's body joints) in multiple frames. For the
> temporal associations, we introduce the Temporal Composite Association Field
> (TCAF) which requires an extended network architecture and training method
> beyond previous Composite Fields. Our experiments show competitive accuracy
> while being an order of magnitude faster on multiple publicly available datasets
> such as COCO, CrowdPose and the PoseTrack 2017 and 2018 datasets. We also show
> that our method generalizes to any class of semantic keypoints such as car and
> animal parts to provide a holistic perception framework that is well suited for
> urban mobility such as self-driving cars and delivery robots.
Previous [CVPR 2019 paper](http://openaccess.thecvf.com/content_CVPR_2019/html/Kreiss_PifPaf_Composite_Fields_for_Human_Pose_Estimation_CVPR_2019_paper.html).
# [Guide](https://openpifpaf.github.io/intro.html)
Detailed documentation is in our __[OpenPifPaf Guide](https://openpifpaf.github.io/intro.html)__.
For developers, there is also the
__[DEV Guide](https://openpifpaf.github.io/dev/intro.html)__
which is the same guide but based on the latest code in the `main` branch.
# Examples
Image credit: "[Learning to surf](https://www.flickr.com/photos/fotologic/6038911779/in/photostream/)" by fotologic which is licensed under [CC-BY-2.0].
Created with:
```sh
pip3 install matplotlib openpifpaf
python3 -m openpifpaf.predict docs/coco/000000081988.jpg --image-output
```
---
Here is the [tutorial for body, foot, face and hand keypoints](https://openpifpaf.github.io/plugins_wholebody.html). Example:
Image credit: [Photo](https://de.wikipedia.org/wiki/Kamil_Vacek#/media/Datei:Kamil_Vacek_20200627.jpg) by [Lokomotive74](https://commons.wikimedia.org/wiki/User:Lokomotive74) which is licensed under [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/).
Created with:
```sh
python -m openpifpaf.predict guide/wholebody/soccer.jpeg \
--checkpoint=shufflenetv2k30-wholebody --line-width=2 --image-output
```
---
Here is the [tutorial for car keypoints](https://openpifpaf.github.io/plugins_apollocar3d.html). Example:
Image credit: [Photo](https://commons.wikimedia.org/wiki/File:Streets_of_Saint_Petersburg,_Russia.jpg) by [Ninaras](https://commons.wikimedia.org/wiki/User:Ninaras) which is licensed under [CC-BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/).
Created with:
```sh
python -m openpifpaf.predict guide/images/peterbourg.jpg \
--checkpoint shufflenetv2k16-apollo-24 -o images \
--instance-threshold 0.05 --seed-threshold 0.05 \
--line-width 4 --font-size 0
```
---
Here is the [tutorial for animal keypoints (dogs, cats, sheep, horses and cows)](https://openpifpaf.github.io/plugins_animalpose.html). Example:
```sh
python -m openpifpaf.predict guide/images tappo_loomo.jpg \
--checkpoint=shufflenetv2k30-animalpose \
--line-width=6 --font-size=6 --white-overlay=0.3 \
--long-edge=500
```
# Commercial License
The open source license is in the [LICENSE](https://github.com/openpifpaf/openpifpaf/blob/main/LICENSE) file.
This software is also available for licensing via the EPFL Technology Transfer
Office (https://tto.epfl.ch/, [email protected]).
[CC-BY-2.0]: https://creativecommons.org/licenses/by/2.0/