This website shares the codes of the "Detail-recovery Image Deraining via Context Aggregation Networks",CVPR 2020.
Prerequisites: tensorflow == 1.10.0 keras == 2.2.4 python == 3.6 CUDA ==10.0
For train the Rain200H, please run: python train.py --image_dir_noise you rain data --image_dir_original you gt data --test_dir_noise you test rain data --test_dir_original you test gt data --If_n True
For train the Rain200L, please run: python train.py --image_dir_noise you rain data --image_dir_original you gt data --test_dir_noise you test rain data --test_dir_original you test gt data --If_n False
For train the Rain800, please run: python train.py --image_dir_noise you rain data --image_dir_original you gt data --test_dir_noise you test rain data --test_dir_original you test gt data --If_n False
The dataset "Rain200H" and "Rain200L" you can download here: https://www.icst.pku.edu.cn/struct/Projects/joint_rain_removal.html
The dataset "Rain800" you can download here: https://drive.google.com/drive/folders/0Bw2e6Q0nQQvGbi1xV1Yxd09rY2s