Clone this repo.
git clone https://github.com/yhjo09/AdaTarget
cd AdaTarget
Download pretrained models and extract it.
unzip models.zip
Download a zip file containing the images of the Gaussian8 test set, and extract it.
unzip Gaussian8.zip
Run.
python test_Iso.py
Output images will be in ./output/Iso
.
Download DIV2KRK test set (from KernelGAN) and unzip the zip file.
unzip DIV2KRK.zip
Run.
python test_RK.py
Output images will be in ./output/RK
.
Download a zip file contains the images of the Set5, Set14, BSDS100, Urban100, and Manga109 test sets, and extract it.
unzip Bicubic.zip
Run.
python test_Bic.py
Output images will be in ./output/Bic
.
Soon
Prepare training images.
./DIV2K/DIV2K_train_HR/*.png
../DIV2K/DIV2K_train_LR_RK/*.png
.Prepare validation images.
Specify the pretrained parameters of the localization network to variable PRETRAINED_LOCNET
.
Start training.
python train_RK.py
./pt_log
../AdaTarget/
.@InProceedings{jo2021adatarget,
author = {Jo, Younghyun and Oh, Seoung Wug and Vajda, Peter and Kim, Seon Joo},
title = {Tackling the Ill-Posedness of Super-Resolution through Adaptive Target Generation},
booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2021}
}