Unofficial implementation of natural and Realistic Single Image Super-Resolution with Explicit Natural Manifold Discrimination (CVPR, 2019) in pytorch (w/ audit-friendly code)

• official tensorflow implementation : https://github.com/JWSoh/NatSR
• paper : CVPR2019

Work In Progress (WIP)

• Python 3.x (recommended 3.7)
• Pytorch 1.x

Recently, many convolutional neural networks for single image super-resolution (SISR) have been proposed, which focus on reconstructing the high-resolution images in terms of objective distortion measures. However, the networks trained with objective loss functions generally fail to reconstruct the realistic fine textures and details that are essential for better perceptual quality. Recovering the realistic details remains a challenging problem, and only a few works have been proposed which aim at increasing the perceptual quality by generating enhanced textures. However, the generated fake details often make undesirable artifacts and the overall image looks somewhat unnatural. Therefore, in this paper, we present a new approach to reconstructing realistic super-resolved images with high perceptual quality, while maintaining the naturalness of the result. In particular, we focus on the domain prior properties of SISR problem. Specifically, we define the naturalness prior in the low-level domain and constrain the output image in the natural manifold, which eventually generates more natural and realistic images. Our results show better naturalness compared to the recent super-resolution algorithms including perception-oriented ones.

• DIV2K : download

0. Clone the repository

$ git clone https://github.com/kozistr/NatSR-pytorch
$ cd ./NatSR-pytorch

1. Configure your own environment.

1.1. Using pipenv by given Pipfile

$ pip3 install -U pipenv
$ pipenv install --dev

1.2. Using requirements.txt

$ pip3 install -r requirements.txt

2. Change the parameter what you want config.yaml

2.1. Mode

• At train : mode: train
• At test : mode: test
• At inference : mode: inference

2.2. Model Type

• training NMD : model_type: 'nmd'
• training FRSR : model_type: 'frsr'

3. Run!

$ python3 -m natsr

@InProceedings{Soh_2019_CVPR,
  author = {Soh, Jae Woong and Park, Gu Yong and Jo, Junho and Cho, Nam Ik},
  title = {Natural and Realistic Single Image Super-Resolution With Explicit Natural Manifold Discrimination},
  booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  month = {June},
  year = {2019}
}

Hyeongchan Kim / @kozistr