Metadata-Version: 2.1
Name: openpifpaf
Version: 0.12a2
Summary: PifPaf: Composite Fields for Human Pose Estimation
Home-page: https://github.com/vita-epfl/openpifpaf
Author: Sven Kreiss
Author-email: research@svenkreiss.com
License: GNU AGPLv3
Platform: UNKNOWN
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: numpy (>=1.16)
Requires-Dist: pysparkling
Requires-Dist: python-json-logger
Requires-Dist: scipy
Requires-Dist: torch (>=1.3.1)
Requires-Dist: torchvision (>=0.4)
Requires-Dist: pillow
Requires-Dist: dataclasses ; python_version < "3.7"
Provides-Extra: coreml
Requires-Dist: coremltools (==4.0b3) ; extra == 'coreml'
Provides-Extra: dev
Requires-Dist: flameprof ; extra == 'dev'
Requires-Dist: jupyter-book (>=0.8.0) ; extra == 'dev'
Requires-Dist: matplotlib ; extra == 'dev'
Requires-Dist: nbdime ; extra == 'dev'
Requires-Dist: nbstripout ; extra == 'dev'
Provides-Extra: onnx
Requires-Dist: onnx ; extra == 'onnx'
Requires-Dist: onnxruntime ; extra == 'onnx'
Requires-Dist: onnx-simplifier (>=0.2.9) ; extra == 'onnx'
Provides-Extra: test
Requires-Dist: nbval ; extra == 'test'
Requires-Dist: onnx ; extra == 'test'
Requires-Dist: onnxruntime ; extra == 'test'
Requires-Dist: onnx-simplifier (>=0.2.9) ; extra == 'test'
Requires-Dist: pylint ; extra == 'test'
Requires-Dist: pytest ; extra == 'test'
Requires-Dist: opencv-python ; extra == 'test'
Requires-Dist: thop ; extra == 'test'
Provides-Extra: train
Requires-Dist: matplotlib ; extra == 'train'
Requires-Dist: pycocotools (>=2.0.1) ; extra == 'train'

# openpifpaf

Continuously tested on Linux, MacOS and Windows: [![Build Status](https://travis-ci.org/vita-epfl/openpifpaf.svg?branch=master)](https://travis-ci.org/vita-epfl/openpifpaf)<br />
[CVPR 2019 paper](http://openaccess.thecvf.com/content_CVPR_2019/html/Kreiss_PifPaf_Composite_Fields_for_Human_Pose_Estimation_CVPR_2019_paper.html)
<!-- [arxiv.org/abs/1903.06593](https://arxiv.org/abs/1903.06593) -->

> PifPaf: Composite Fields for Human Pose Estimation
>
> We propose a new bottom-up method for multi-person 2D human pose
> estimation that is particularly well suited for urban mobility such as self-driving cars
> and delivery robots. The new method, PifPaf, uses a Part Intensity Field (PIF) to
> localize body parts and a Part Association Field (PAF) to associate body parts with each other to form
> full human poses.
> Our method outperforms previous methods at low resolution and in crowded,
> cluttered and occluded scenes
> thanks to (i) our new composite field PAF encoding fine-grained information and (ii) the choice of Laplace loss for regressions which incorporates a notion of uncertainty.
> Our architecture is based on a fully
> convolutional, single-shot, box-free design.
> We perform on par with the existing
> state-of-the-art bottom-up method on the standard COCO keypoint task
> and produce state-of-the-art results on a modified COCO keypoint task for
> the transportation domain.


# Example

![example image with overlaid pose predictions](https://github.com/vita-epfl/openpifpaf/raw/master/docs/coco/000000081988.jpg.predictions.png)

Image credit: "[Learning to surf](https://www.flickr.com/photos/fotologic/6038911779/in/photostream/)" by fotologic which is licensed under [CC-BY-2.0].<br />
Created with:
```sh
pip3 install openpifpaf matplotlib
python3 -m openpifpaf.predict coco/000000081988.jpg --image-output
```


# [Guide](https://vita-epfl.github.io/openpifpaf/intro.html)

Continue to our __[OpenPifPaf Guide](https://vita-epfl.github.io/openpifpaf/intro.html)__.


[CC-BY-2.0]: https://creativecommons.org/licenses/by/2.0/


# Commercial License

This software is available for licensing via the EPFL Technology Transfer
Office (https://tto.epfl.ch/, info.tto@epfl.ch).


