SCGAN-TensorFlow

An implementation of SCGAN using Torch. You can click here to visit Tensorflow version.

Original paper: https://arxiv.org/abs/2210.07594

Results on test data

haze -> clear

Input	Output

Prerequisites

Linux or OSX
NVIDIA GPU + CUDA CuDNN (CPU mode and CUDA without CuDNN may work with minimal modification, but untested)
For MAC users, you need the Linux/GNU commands gfind and gwc, which can be installed with brew install findutils coreutils.

Getting Started

Installation

Install torch and dependencies from https://github.com/torch/distro
Install torch packages nngraph, class, display

luarocks install nngraph
luarocks install class
luarocks install https://raw.githubusercontent.com/szym/display/master/display-scm-0.rockspec

Clone this repo:

git clone https://github.com/junyanz/CycleGAN
cd CycleGAN

Data preparing

First, download the dataset

Unpaired dataset: The dataset is built by ourselves, and there are all real haze images from website.

10000 images: Address：Baidu cloud disk Extraction code：zvh6

1000 images: Address：Baidu cloud disk Extraction code:47v9
Paired dataset: The dataset is added haze by ourselves according to the image depth.

Address: Baidu cloud disk Extraction code : 63xf

Now, let's generate dehaze images:

DATA_ROOT=./datasets/test name=dehaze_pretrained model=one_direction_test phase=test loadSize=256 fineSize=256 resize_or_crop="scale_width" th test.lua

The test results will be saved to ./results/dehaze_pretrained/latest_test/index.html.

Train

Download a dataset (e.g. zebra and horse images from ImageNet):

bash ./datasets/download_dataset.sh haze2dehaze

Train a model:

DATA_ROOT=./datasets/haze2dehaze name=haze2dehaze_model th train.lua

(CPU only) The same training command without using a GPU or CUDNN. Setting the environment variables gpu=0 cudnn=0 forces CPU only

DATA_ROOT=./datasets/dehaze2haze name=haze2dehaze_model gpu=0 cudnn=0 th train.lua

(Optionally) start the display server to view results as the model trains. (See Display UI for more details):

th -ldisplay.start 8000 0.0.0.0

Test

Finally, test the model:

DATA_ROOT=./datasets/haze2dehaze name=haze2dehaze_model phase=test th test.lua

The test results will be saved to a html file here: ./results/haze2dehaze_model/latest_test/index.html.

Model Zoo

Download the pre-trained models with the following script. The model will be saved to ./checkpoints/model_name/latest_net_G.t7.

Display UI

Optionally, for displaying images during training and test, use the display package.

Install it with: luarocks install https://raw.githubusercontent.com/szym/display/master/display-scm-0.rockspec
Then start the server with: th -ldisplay.start
Open this URL in your browser: http://localhost:8000

By default, the server listens on localhost. Pass 0.0.0.0 to allow external connections on any interface:

th -ldisplay.start 8000 0.0.0.0

Then open http://(hostname):(port)/ in your browser to load the remote desktop.

Citation

If you use this code for your research, please cite our paper:

@article{zhang2022see,
  title={See Blue Sky: Deep Image Dehaze Using Paired and Unpaired Training Images},
  author={Zhang, Xiaoyan and Tang, Gaoyang and Zhu, Yingying and Tian, Qi},
  journal={arXiv preprint arXiv:2210.07594},
  year={2022}
}

Related Projects:

CycleGAN pix2pix

Acknowledgments

Code borrows from CycleGAN.

tygrer/SCGAN-torch