ESRGAN

ESRGAN

This is the implementation of "Enhanced Super-Resolution Generative Adversarial Network".
Original paper: X. Wang, K. Yu, S. Wu, J. Gu, Y. Liu, C. Dong, Y. Qiao, and C. C. Loy. ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks. In Proceedings of the European Conference on Computer Vision, 2018. link

Usage

0. Download pre-trained model

Please download VGG19 pre-trained model with ImageNet.

$ wget https://huggingface.co/koba-jon/pre-train_cpp/resolve/main/models/vgg19_bn.pth

1. Build

Please build the source file according to the procedure.

$ mkdir build
$ cd build
$ cmake ..
$ make -j4
$ cd ..

2. Dataset Setting

Recommendation

• Large-scale CelebFaces Attributes (CelebA) Dataset
  This is a large-scale face attributes dataset with more than 200K celebrity images, each with 40 attribute annotations.
  Link: official

Setting

Please create a link for the dataset.
The following hierarchical relationships are recommended.

datasets
|--Dataset1
|    |--trainLR
|    |    |--image1.png
|    |    |--image2.bmp
|    |    |--image3.jpg
|    |
|    |--trainHR
|    |    |--image1.png
|    |    |--image2.bmp
|    |    |--image3.jpg
|    |
|    |--validLR
|    |--validHR
|    |--testLR
|    |--testHR
|
|--Dataset2
|--Dataset3

You should substitute the path of training low-resolution image for "<training_lr_path>", training high-resolution image for "<training_hr_path>", test low-resolution image for "<test_lr_path>", test high-resolution image for "<test_hr_path>", respectively.
The following is an example for "celebA".

$ cd datasets
$ mkdir celebA
$ cd celebA
$ ln -s <training_lr_path> ./trainLR
$ ln -s <training_hr_path> ./trainHR
$ ln -s <test_lr_path> ./testLR
$ ln -s <test_hr_path> ./testHR
$ cd ../..

3. Training

Setting

Please set the shell for executable file.

$ vi scripts/train.sh

The following is an example of the training phase.
If you want to view specific examples of command line arguments, please view "src/main.cpp" or add "--help" to the argument.

#!/bin/bash

DATA='celebA'

./ESRGAN \
    --train true \
    --epochs 300 \
    --dataset ${DATA} \
    --hr_size 256 \
    --batch_size 16 \
    --gpu_id 0 \
    --nc 3

Run

Please execute the following to start the program.

$ sh scripts/train.sh

4. Test

Setting

Please set the shell for executable file.

$ vi scripts/test.sh

The following is an example of the test phase.
If you want to view specific examples of command line arguments, please view "src/main.cpp" or add "--help" to the argument.

#!/bin/bash

DATA='celebA'

./ESRGAN \
    --test true \
    --dataset ${DATA} \
    --hr_size 256 \
    --gpu_id 0 \
    --nc 3

There are no particular restrictions on both input and output images.
However, the two file names must correspond without the extension.

Run

Please execute the following to start the program.

$ sh scripts/test.sh