Depth-Attentional Features for Single-Image Rain Removal

by Xiaowei Hu, Chi-Wing Fu, Lei Zhu, and Pheng-Ann Heng

This implementation is written by Xiaowei Hu at the Chinese University of Hong Kong.

---

Please find the code of the journal version at https://github.com/xw-hu/DGNL-Net/.

---

RainCityscapes Dataset

Our RainCityscapes dataset is available for download at the Cityscapes website.

Citations

@InProceedings{Hu_2019_CVPR,
     author = {Hu, Xiaowei and Fu, Chi-Wing and Zhu, Lei and Heng, Pheng-Ann},
     title = {Depth-Attentional Features for Single-Image Rain Removal},
     booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
     pages={8022--8031},
     year = {2019}
}

@article{hu2020single,
     author = {Hu, Xiaowei and Zhu, Lei and Wang, Tianyu and Fu, Chi-Wing and Heng, Pheng-Ann},
     title = {Single-Image Real-Time Rain Removal Based on Depth-Guided Non-Local Features},
     journal = {IEEE Transactions on Image Processing},
     year = {2020},
     note = {to appear}
}

Installation

1. Please download and compile our CF-Caffe.

2. link the CF-Caffe to DAF-Net/caffe.

   ln -s '/path/to/CF-Caffe' '/path/to/DAF-Net/caffe'

Train

Download the pre-trained VGG16 model at http://www.robots.ox.ac.uk/~vgg/research/very_deep/.
Put this model in CF-Caffe/models/

1. Enter the ./examples/DAF-Net/
   Modify the image path in train_val.prototxt.

2. Run

   sh train.sh

Test

1. Please download our pretrained model at Google Drive.
   Put this model in ./examples/DAF-Net/snapshot/.

2. Enter the ./examples/ and run test_raincityscapes.m in Matlab.

Evaluation

Enter the DAF-Net/examples/ and run evaluate_raincityscapes.m in Matlab.