README.md

July 11, 2023 ยท View on GitHub

Practical Single-Image Super-Resolution Using Look-Up Table

[Paper]

This repository contains an op-for-op Paddle reimplementation of Practical Single-Image Super-Resolution Using Look-Up Table. It is modified from the original source code of Pytorch implementation(https://github.com/yhjo09/SR-LUT)

Dependency

  • Python 3.7
  • Paddle
  • glob
  • numpy
  • pillow
  • tqdm
  • tensorboardx

1.Dataset

We use DIV2K dataset for training and Set5 dataset for testing

2. Testing

  • The testing image pairs is placed in ./val folder,you can change the image address in line 24 in val_model_F.py
  • The result image will be put in ./result/F/HR/
  • python val_model_F.py

BibTeX

@InProceedings{jo2021practical,
   author = {Jo, Younghyun and Kim, Seon Joo},
   title = {Practical Single-Image Super-Resolution Using Look-Up Table},
   booktitle = {The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
   month = {June},
   year = {2021}
}