M2FP: Mask2Former for Parsing

Deep Learning Technique for Human Parsing: A Survey and Outlook

If you find this repository helpful, please consider citing:

@article{yang2023humanparsing,
  title={Deep Learning Technique for Human Parsing: A Survey and Outlook},
  author={Lu Yang and Wenhe Jia and Shan Li and Qing Song},
  journal={arXiv preprint arXiv:2301.00394},
  year={2023}
}

Features

A single architecture for single human parsing, and multiple (instance-level) human parsing.
Support several parsing datasets: LIP, PASCAL-Person-Part, CIHP, MHP-v2.

Updates

[2023/1/19] models in GoogleDrive are released.

[2023/1/3] paper and code released.

[2022/6/19] code initialization.

Installation

See installation instructions.

Getting Started

See Preparing Datasets for M2FP.

See Getting Started with M2FP.

Results and Models

Datasets	mIoU	APr	APp	DOWNLOAD
LIP	59.86			BaiduCloud (passwd: 36ec), GoogleDrive
PASCAL-Person-Part	72.54	56.46
CIHP	69.15	60.47		BaiduCloud (passwd: jzrn), GoogleDrive
MHP-v2	47.64		53.36	BaiduCloud (passwd: seel), GoogleDrive

License

Shield:

The majority of M2FP is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

Citing MaskFormer and Mask2Former

If you find the code useful, please also consider the following MaskFormer and Mask2Former BibTeX entry.

@inproceedings{cheng2021mask2former,
  title={Masked-attention Mask Transformer for Universal Image Segmentation},
  author={Bowen Cheng and Ishan Misra and Alexander G. Schwing and Alexander Kirillov and Rohit Girdhar},
  journal={CVPR},
  year={2021}
}

@inproceedings{cheng2021maskformer,
  title={Per-Pixel Classification is Not All You Need for Semantic Segmentation},
  author={Bowen Cheng and Alexander G. Schwing and Alexander Kirillov},
  journal={NeurIPS},
  year={2021}
}

Acknowledgement

Code is largely based on Mask2Former (https://github.com/facebookresearch/Mask2Former).

soeaver/M2FP