Investigating Loss Functions for Extreme Super-Resolution

Younghyun Jo, Sejong Yang, Seon Joo Kim; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2020, pp. 424-425

Abstract


The performance of image super-resolution (SR) has been greatly improved by using convolutional neural networks. Most of the previous SR methods have been studied up to x4 upsampling, and few were studied for x16 upsampling. The general approach for perceptual x4 SR is using GAN with VGG based perceptual loss, however, we found that it creates inconsistent details for perceptual x16 SR. To this end, we have investigated loss functions and we propose to use GAN with LPIPS loss for perceptual extreme SR. In addition, we use U-net structure discriminator together to consider both the global and local context of an input image. Experimental results show that our method outperforms the conventional perceptual loss, and we achieved second place in preliminary results of NTIRE 2020 perceptual extreme SR challenge.

Related Material


[pdf]
[bibtex]
@InProceedings{Jo_2020_CVPR_Workshops,
author = {Jo, Younghyun and Yang, Sejong and Kim, Seon Joo},
title = {Investigating Loss Functions for Extreme Super-Resolution},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2020}
}