SRFeat: Single Image Super-Resolution with Feature Discrimination

Seong-Jin Park
POSTECH

Hyeongseok Son
POSTECH

Sunghyun Cho
DGIST

Ki-Sang Hong
POSTECH

Seungyong Lee
POSTECH

European Conference on Computer Vision (ECCV), September 2018




Abstract

Generative adversarial networks (GANs) have recently been adopted to single image super-resolution (SISR) and showed impressive results with realistically synthesized high-frequency textures. However, the results of such GAN-based approaches tend to include less meaningful high-frequency noise that is irrelevant to the input image. In this paper, we propose a novel GAN-based SISR method that overcomes the limitation and produces more realistic results by attaching an additional discriminator that works in the feature domain. Our additional discriminator encourages the generator to produce structural high-frequency features rather than noisy artifacts as it distinguishes synthetic and real images in terms of features. We also design a new generator that utilizes long-range skip connections so that information between distant layers can be transferred more effectively. Experiments show that our method achieves the state-of-the-art performance in terms of both PSNR and perceptual quality compared to recent GAN-based methods.


Paper
PDF, 8.73MB
Code
Github Link
see the new features in coupe website [link]

BibTex

@InProceedings{Park_2018_ECCV,
author = {Park, Seong-Jin and Son, Hyeongseok and Cho, Sunghyun and Hong, Ki-Sang and Lee, Seungyong},
title = {SRFeat: Single Image Super-Resolution with Feature Discrimination},
booktitle = {European Conference on Computer Vision (ECCV)},
month = {September},
year = {2018}
}