Image Super-Resolution by Learning Weighted Convolutional Sparse Coding

This repository is for LW-CSC.

Dependencies

• Python 2 (Recommend to use Anaconda)
• Pytorch 1.0.1
• NVIDIA GPU + CUDA
• Python packages: pip install xxx

Train

Prepare training data

1. Download the 291 images (Baidu Netdisk psw:ryjr), and place them in './data' folder.
2. cd to './data', and run generate_train.m to generate training data.

Begin to train

1. (optional) Download the model for our paper and place it in './pretrained'.

2. Run the following script to train.

   bash train.sh

Test

1. Run the following script to evaluate.

   python evaluate.py

Citation

If you use any part of this code in your research, please cite our paper:

@article{lwcsc2021,
  title={Image Super-Resolution by Learning Weighted Convolutional Sparse Coding},
  author={He, Jingwei and Yu, Lei and Liu, Zhou and Yang, Wen},
  journal={Signal, Image and Video Processing},
  volume={x},
  number={x},
  pages={xx--xx},
  year={2021},
  publisher={Springer}
}