Densely-Self-guided-Wavelet-Network-for-Image-Denoising Official PyTorch Implementation

Configuration Environment

Our network uses DWT and IDWT. Please install correspinding library as the following link: https://github.com/fbcotter/pytorch_wavelets

We have upload all the .py files and .txt file. Please unzip the training and valid data in the workspace as name_list.txt and val_gt.txt.

Datasets and pre-trained networks

Download the pre-trained model Google drive

Download testing data Google drive

Testing

python submit.py

Please set parser.add_argument.use_ensemble as True when you test our model. The code will generate our ensemble results.

If you need to test the runtime, please change the parser.add_argument.use_ensemble in submit.py as False.

Training

python train.py

Citation

If you find this work useful for your research, please cite our paper:

@inproceedings{liu2020densely,
  title={Densely self-guided wavelet network for image denoising},
  author={Liu, Wei and Yan, Qiong and Zhao, Yuzhi},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  pages={432--433},
  year={2020}
}

Reference

[1] Liu Wei，Yan Qiong，Zhao Yuzhi. Densely Self-guided Wavelet Network for Image Denoising[C]. IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops 2020 (CVPRW)

[2] S. Gu, Y. Li, L. V. Gool, and R. Timofte, "Self-Guided Network for Fast Image Denoising”

[3] P. Liu, H. Zhang, W. Lian, and W. Zuo, "Multi-level Wavelet Convolutional Neural Networks."