DAWN: Direction-aware Attention Wavelet Network for Image Deraining (ACM MM2023)
Kui Jiang, Wenxuan Liu, Zheng Wang, Xian Zhong, Junjun Jiang and Chia-Wen Lin
Paper: DAWN: Direction-aware Attention Wavelet Network for Image Deraining
The model is built in PyTorch 1.1.0 and tested on Ubuntu 16.04 environment (Python3.7, CUDA9.0, cuDNN7.5).
For installing, follow these intructions
conda create -n pytorch1 python=3.7
conda activate pytorch1
conda install pytorch=1.1 torchvision=0.3 cudatoolkit=9.0 -c pytorch
pip install matplotlib scikit-image opencv-python yacs joblib natsort h5py tqdm
To test the pre-trained deraining model on your own images, run
python test.py
-
Download the Datasets
-
Train the model with default arguments by running
python train.py
-
Download the model and place it in
./pretrained_models/ -
Download test datasets (Test100, Rain100H, Rain100L, Test1200, Test2800) from here and place them in
./Datasets/Synthetic_Rain_Datasets/test/ -
Run
python test.py
evaluate_PSNR_SSIM.m
Experiments are performed for different image processing tasks including, image deraining, image dehazing and low-light image enhancement.
Code borrows from MPRNet by Syed Waqas Zamir. Thanks for sharing !
If you use DAWN, please consider citing:
@inproceedings{jiang2023dawn, title={Dawn: Direction-aware attention wavelet network for image deraining}, author={Jiang, Kui and Liu, Wenxuan and Wang, Zheng and Zhong, Xian and Jiang, Junjun and Lin, Chia-Wen}, booktitle={Proceedings of the 31st ACM International Conference on Multimedia}, pages={7065--7074}, year={2023} }
Should you have any question, please contact Kui Jiang ([email protected])