/ Machine Learning

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Super Resolution Examples

πŸš€πŸš€πŸš€πŸš€πŸš€πŸš€ THIS PROJECT WILL BE CLOSED AND MOVED TO THIS FOLDER IN A MONTH.

πŸš€πŸš€πŸš€πŸš€πŸš€πŸš€ THIS PROJECT WILL BE CLOSED AND MOVED TO THIS FOLDER IN A MONTH.

πŸš€πŸš€πŸš€πŸš€πŸš€πŸš€ THIS PROJECT WILL BE CLOSED AND MOVED TO THIS FOLDER IN A MONTH.

SRGAN Architecture

TensorFlow Implementation of "Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network"

Results

SRGAN_Result2

SRGAN_Result3

Prepare Data and Pre-trained VGG

    1. You need to download the pretrained VGG19 model in here as tutorial_models_vgg19.py show.
    1. You need to have the high resolution images for training.
    • In this experiment, I used images from DIV2K - bicubic downscaling x4 competition, so the hyper-paremeters in config.py (like number of epochs) are seleted basic on that dataset, if you change a larger dataset you can reduce the number of epochs.
    • If you dont want to use DIV2K dataset, you can also use Yahoo MirFlickr25k, just simply download it using train_hr_imgs = tl.files.load_flickr25k_dataset(tag=None) in main.py.
    • If you want to use your own images, you can set the path to your image folder via config.TRAIN.hr_img_path in config.py.

Run

config.TRAIN.img_path = "your_image_folder/"
  • Start training.
python train.py
  • Start evaluation.
python train.py --mode=evaluate 

Reference

Author

Citation

If you find this project useful, we would be grateful if you cite the TensorLayer paper:

@article{tensorlayer2017,
author = {Dong, Hao and Supratak, Akara and Mai, Luo and Liu, Fangde and Oehmichen, Axel and Yu, Simiao and Guo, Yike},
journal = {ACM Multimedia},
title = {{TensorLayer: A Versatile Library for Efficient Deep Learning Development}},
url = {http://tensorlayer.org},
year = {2017}
}

Other Projects

Discussion

GitHub

Comments