Metadata-Version: 2.1
Name: panoptic_parts
Version: 2.0rc5
Summary: Panoptic Parts datasets
Home-page: https://github.com/pmeletis/panoptic_parts
Author: Panoptic Parts team
Author-email: panoptic.parts@outlook.com
Maintainer: Panagiotis Meletis
Maintainer-email: panoptic.parts@outlook.com
License: UNKNOWN
Project-URL: Documentation, https://panoptic-parts.readthedocs.io
Project-URL: Bug Tracker, https://github.com/pmeletis/panoptic_parts/issues
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.7
Classifier: Operating System :: OS Independent
Requires-Python: >=3.7
Description-Content-Type: text/markdown
Provides-Extra: MERGING
License-File: LICENSE

# Part-aware Panoptic Segmentation

[![Documentation Status](https://readthedocs.org/projects/panoptic-parts/badge/?version=stable)](https://panoptic-parts.readthedocs.io/en/stable/?badge=stable)

**v2.0 Release Candidate**

## [[CVPR 2021 Paper](https://openaccess.thecvf.com/content/CVPR2021/html/de_Geus_Part-Aware_Panoptic_Segmentation_CVPR_2021_paper.html)] [[Datasets Technical Report](https://arxiv.org/abs/2004.07944 "arxiv.org")] [[Documentation](https://panoptic-parts.readthedocs.io/en/stable)]

This repository contains code and tools for reading, processing, evaluating on, and visualizing Panoptic Parts datasets. Moreover, it contains code for reproducing our CVPR 2021 paper results.

## Datasets

*Cityscapes-Panoptic-Parts* and *PASCAL-Panoptic-Parts* are created by extending two established datasets for image scene understanding, namely [Cityscapes](https://github.com/mcordts/cityscapesScripts "Cityscapes") and [PASCAL](http://host.robots.ox.ac.uk/pascal/VOC/voc2010/ "PASCAL") datasets. Detailed description of the datasets and various statistics are presented in our technical report in [arxiv](https://arxiv.org/abs/2004.07944 "arxiv.org"). The datasets can be downloaded from:

- [Cityscapes Panoptic Parts](https://www.cityscapes-dataset.com/login/)
- [PASCAL Panoptic Parts](https://1drv.ms/u/s!AojlpuGgPtL1bHXfIdeL14IeVhI?e=5tNfET) ([alternative link](https://pan.baidu.com/s/1k96Wdg_IyD91kvq87Wy7nw), code: i7ap)

### Examples

![Image](readme/aachen_000012_000019_leftImg8bit.jpg "Image") | ![Image](readme/aachen_000012_000019_uids_pids_colored.png "Image")
---- | ----
![Image](readme/2008_000393.jpg "Image") | ![Image](readme/2008_000393_colored.png "Image") | ![Image](readme/2008_000716.jpg "Image") | 

More examples [here](https://panoptic-parts.readthedocs.io/en/stable/visualization.html).

## Installation

The code can be used by installing the requirements and cloning the repository (at least Python 3.7 is needed). A pip package will be available soon.

```shell
git clone https://github.com/pmeletis/panoptic_parts.git
cd panoptic_parts
pip install -r requirements.txt
```

## API and code reference

We provide a public, stable API, and various code utilities that are documented [here](https://panoptic-parts.readthedocs.io/en/stable/api_and_code.html).


## Reproducing CVPR 2021 paper

The part-aware panoptic segmentation results from the paper can be reproduced using [this](https://panoptic-parts.readthedocs.io/en/stable/generate_results.html) guide.

## Evaluation metrics

We provide two metrics for evaluating performance on Panoptic Parts datasets.

- Part-aware Panoptic Quality (PartPQ): [here](https://panoptic-parts.readthedocs.io/en/stable/evaluate_results.html).
- Intersection over Union (IoU): _TBA_

## Citations

 Please cite us if you find our work useful or you use it in your research:

```bibtex
@inproceedings{degeus2021panopticparts,
    title = {Part-aware Panoptic Segmentation},
    author = {Daan de Geus and Panagiotis Meletis and Chenyang Lu and Xiaoxiao Wen and Gijs Dubbelman},
    booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    year = {2021}
}
```

```bibtex
@article{meletis2020panopticparts,
    title = {Cityscapes-Panoptic-Parts and PASCAL-Panoptic-Parts datasets for Scene Understanding},
    author = {Panagiotis Meletis and Xiaoxiao Wen and Chenyang Lu and Daan de Geus and Gijs Dubbelman},
    type = {Technical report},
    institution = {Eindhoven University of Technology},
    date = {16/04/2020},
    url = {https://github.com/tue-mps/panoptic_parts},
    eprint={2004.07944},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}
```

<a href="https://www.tue.nl/en/research/research-groups/signal-processing-systems/mobile-perception-systems-lab"><img src="docs/source/_static/mps_logo.png" height="100" alt="MPS"></a> &emsp; <a href="https://www.tue.nl"><img src="docs/source/_static/tue_logo.jpg" height="100" alt="TU/e"></a>

## Contact

Please feel free to contact us for any suggestions or questions.

**panoptic.parts@outlook.com**

The Panoptic Parts datasets team

Correspondence: Panagiotis Meletis, Vincent (Xiaoxiao) Wen


