jisus189 / C3Net

C3Net: Demoireing Network Attentive in Channel, Color and Concatenation (CVPRW 2020)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

C3Net

This is a PyTorch implementation of the New Trends in Image Restoration and Enhancement workshop and challenges on image and video restoration and enhancement (NTIRE 2020 with CVPR 2020) paper, C3Net: Demoireing Network Attentive in Channel, Color and Concatenation.

If you find our project useful in your research, please consider citing:

@InProceedings{Kim_2020_CVPR_Workshops,
author = {Kim, Sangmin and Nam, Hyungjoon and Kim, Jisu and Jeong, Jechang},
title = {C3Net: Demoireing Network Attentive in Channel, Color and Concatenation},
booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2020}
}

Dependencies

Python 3.6.9
PyTorch 1.4.0

Data

Reference

Proposed algorithm

C3Net (Track 1: Single Image)
AVC_Block
AttBlock
ResBlock
C3Net-Burst (Track 2: Burst)
AVC_Block-Burst

Training

Use the following command to use our training codes

python train.py

For training pre-trained model, download the model first.
trained model (Track 1: Single Image)
trained model (Track 2: Burst)
Then, set the option --resume to where the downloaded model is.
There are other options you can choose. Please refer to train.py.

Test

Use the following command to use our test codes

python test.py

For testing pre-trained model, download the model first.
trained model (Track 1: Single Image)
trained model (Track 2: Burst)
Then, set the option --logdir to where the downloded model is.
There are other options you can choose. Please refer to test.py.

Results (PSNR/SSIM)

Track 1: Single Image - 41.30/0.99
Track 2: Burst - 40.55/0.99

Contact

If you have any question about the code or paper, feel free to ask me to ksmh1652@gmail.com.

Acknowledgement

Thanks for SaoYan who gave the implementaion of DnCNN.

About

C3Net: Demoireing Network Attentive in Channel, Color and Concatenation (CVPRW 2020)


Languages

Language:Python 100.0%