Residual-SE-Network
Residual Squeeze-and-Excitation Network for Fast Image Deraining
Jun Fu, Jianfeng Xu, Kazuyuki Tasaka, Zhibo Chen
https://arxiv.org/abs/2006.00757
Modified and add BASNet to suit reflection removal.
Requirements:(All network reimplements are same of similar)
- 1.Pytorch 1.3.0
- 2.Torchvision 0.2.0
- 3.Python 3.6.10
- 4.glob (Dataset)
- 5.PIL
- 6.tqdm(For training)
- 7.Opencv-Python
- 8.tensorboardX
Dataset Modified:
Line 25,26,27
imgpath='/public/zebanghe2/cycledomain/dataset/mix'
transpath='/public/zebanghe2/cycledomain/dataset/transmission'
maskpath='/public/zebanghe2/cycledomain/dataset/mask'
In this project,maskpath is invalid if you just use ResSENet.
Train
python train.py
Epoch Number:Line 96
Batch Size:Line 91
Test python test.py
If any problem, please ask in issue.
Jerry He