git《Investigating Loss Functions for Extreme Super-Resolution》(CVPR 2020) GitHub:

Investigating Loss Functions for Extreme Super-Resolution

NTIRE 2020 Perceptual Extreme Super-Resolution Submission.

Our method ranked first and second in PI and LPIPS measures respectively.

[Paper]

Dependency

• Python 3.6
• PyTorch 1.2
• numpy
• pillow
• tqdm

Test

1. Clone this repo.

git clone https://github.com/kingsj0405/ciplab-NTIRE-2020

2. Download pre-trained model and place it to ./model.pth.

• NTIRE submission version
• Updated version

3. Place low-resolution input images to ./input.

4. Run.

python test.py

If your GPU memory lacks, please try with option -n 3 or a larger number.

5. Check your results in ./output.

Train

1. Clone this repo.

git clone https://github.com/kingsj0405/ciplab-NTIRE-2020

2. Prepare training png images into ./train.

3. Prepare validation png images into ./val.

4. Open train.py and modify user parameters in L22.

5. Run.

python train.py

If your GPU memory lacks, please try with lower batch size or patch size.

BibTeX

@InProceedings{jo2020investigating,
   author = {Jo, Younghyun and Yang, Sejong and Joo Kim, Seon},
   title = {Investigating Loss Functions for Extreme Super-Resolution},
   booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
   month = {June},
   year = {2020}
}

External codes from

• LPIPS
• RRDB (ESRGAN)