Pyramid Attention for Image Restoration

This repository is for PANet and PA-EDSR introduced in the following paper

Yiqun Mei, Yuchen Fan, Yulun Zhang, Jiahui Yu, Yuqian Zhou, Ding Liu, Yun Fu, Thomas S. Huang and Honghui Shi "Pyramid Attention for Image Restoration", [Arxiv]

The code is built on EDSR (PyTorch) & RNAN and tested on Ubuntu 18.04 environment (Python3.6, PyTorch_1.1) with Titan X/1080Ti/V100 GPUs.

Train
Test
Results
Citation
Acknowledgements

Train

Prepare training data

Download DIV2K training data (800 training + 100 validtion images) from DIV2K dataset or SNU_CVLab.
Specify '--dir_data' in optional.py based on the HR and LR images path.
Organize training data like:

DIV2K/
├── DIV2K_train_HR
├── DIV2K_train_LR_bicubic
│   └── X10
│   └── X30
│   └── X50
│   └── X70
├── DIV2K_valid_HR
└── DIV2K_valid_LR_bicubic
    └── X10
    └── X30
    └── X50
    └── X70

For more informaiton, please refer to EDSR(PyTorch).

Begin to train

(optional) All the pretrained models and visual results can be downloaded from Google Drive.

Cd to 'PANet-PyTorch/[Task]/code', run the following scripts to train models.

You can use scripts in file 'demo.sb' to train and test models for our paper.

# Example Usage: N=10
python main.py --n_GPUs 1 --lr 1e-4  --batch_size 16  --n_resblocks 80 --save_models --epoch 1000 --decay 200-400-600-800 --model PANET --scale 50 --patch_size 48 --reset --save PANET_N50 --n_feats 64 --data_train DIV2K --chop

Test

Quick start

Cd to 'PANet-PyTorch/[Task]/code', run the following scripts.

You can use scripts in file 'demo.sb' to produce results for our paper.

# No self-ensemble, use different testsets to reproduce the results in the paper.
# Example Usage: 
python main.py --model PANET --n_resblocks 80 --n_feats 64 --data_test Urban100 --scale 10 --save_results --chop --test_only  --pre_train ../path_to_model

The whole test pipeline

Prepare test data. Organize training data like:

benchmark/
├── testset1
│   └── HR 
│   └── LR_bicubic
│   	└── X10
│   	└── ..
├── testset2

Conduct image denoise.

See Quick start
Evaluate the results.

Run 'Evaluate_PSNR_SSIM.m' to obtain PSNR/SSIM values for paper.

Citation

If you find the code helpful in your resarch or work, please cite the following papers.

@article{mei2020pyramid,
  title={Pyramid Attention Networks for Image Restoration},
  author={Mei, Yiqun and Fan, Yuchen and Zhang, Yulun and Yu, Jiahui and Zhou, Yuqian and Liu, Ding and Fu, Yun and Huang, Thomas S and Shi, Honghui},
  journal={arXiv preprint arXiv:2004.13824},
  year={2020}
}
@InProceedings{Lim_2017_CVPR_Workshops,
  author = {Lim, Bee and Son, Sanghyun and Kim, Heewon and Nah, Seungjun and Lee, Kyoung Mu},
  title = {Enhanced Deep Residual Networks for Single Image Super-Resolution},
  booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
  month = {July},
  year = {2017}
}

Acknowledgements

This code is built on EDSR (PyTorch), RNAN and generative-inpainting-pytorch. We thank the authors for sharing their codes.

hyliu
/

lambdanet