we propose SEIDNet equipped with the generative Status Estimation and Information Decoupling for rain removal.
- python 3.7.11
- pytorch 1.8.0
- CUDA Version 11.4
- opencv-python 4.4.0.44
- scikit-image 0.17.2
Single image rain removal
- Rain100L-old_version https://github.com/nnUyi/DerainZoo/blob/master/DerainDatasets.md
- Rain100H-old_version https://github.com/nnUyi/DerainZoo/blob/master/DerainDatasets.md
- Rain1400 https://xueyangfu.github.io/projects/cvpr2017.html
- SPA https://stevewongv.github.io/derain-project.html
- Rain13k https://github.com/swz30/MPRNet
Single image snow removal
- Snow100K https://goo.gl/BrRc3U
Single image haze removal
Single image shadow removal
Coming soon
sh train_SEIDNet.sh
sh test_SEIDNet.sh
Single image rain removal
We compare SEIDNet with state-of-the-art methods on the test sets of Rain100H, Rain100L, Rain1400, SPA and Rain13K. The performances are reported in terms of PSNR and SSIM.
We compare SEIDNet with state-of-the-art methods on the test sets of Rain13K. The performances are reported in terms of PSNR and SSIM.
Single image snow removal
We compare SEIDNet with state-of-the-art methods on the test sets of Snow100K. The performances are reported in terms of PSNR and SSIM.
Single image haze removal
We compare SEIDNet with other methods on ITS&OTS. The results are listed in terms of PSNR and SSIM.