feiyungu / DFCRN-for-Image-Dehazing

Implementation code of the paper "Deep Fully Convolutional Regression Networks for Single Image Haze Removal"

DFCRN-for-Image-Dehazing

Implementation code of the paper "Deep Fully Convolutional Regression Networks for Single Image Haze Removal."

• Paper link , Microsoft OneDrive: Poster Link and Supplementary_material Link

1. Prerequisites:

• Windows Caffe or Linux Caffe
• MATLAB

2. Installation

• please install Caffe (Windows or Linux) fisrtly, and the installation for Windows OS is here.

3. Demo using pre-trained model

• download the DFCRN_Code folder and run the dfcrn_demo.m on MATLAB.

4. Dehazing results

5. Paper BibTeX

@inproceedings{zhao2017deep,
  title={Deep fully convolutional regression networks for single image haze removal},
  author={Zhao, Xi and Wang, Keyan and Li, Yunsong and Li, Jiaojiao},
  booktitle={Visual Communications and Image Processing (VCIP), 2017 IEEE},
  pages={1--4},
  year={2017},
  organization={IEEE}
}

If you encounter any issue when using our code or model, please let me know. (Email:[e-mail])