EnhanceNet: Single Image Super-Resolution Through Automated Texture Synthesis

Single image super-resolution is the task of inferring a high-resolution image from a single low-resolution input. Traditionally, the performance of algorithms for this task is measured using pixel-wise reconstruction measures such as peak signal-to-noise ratio (PSNR) which have been shown to correl...

Full description

Saved in:
Bibliographic Details
Main Authors: Sajjadi, Mehdi S. M., Scholkopf, Bernhard, Hirsch, Michael
Format: Conference Proceeding
Language:English
Subjects:
Convolution
Image resolution
Interpolation
Signal resolution
Training
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Single image super-resolution is the task of inferring a high-resolution image from a single low-resolution input. Traditionally, the performance of algorithms for this task is measured using pixel-wise reconstruction measures such as peak signal-to-noise ratio (PSNR) which have been shown to correlate poorly with the human perception of image quality. As a result, algorithms minimizing these metrics tend to produce over-smoothed images that lack highfrequency textures and do not look natural despite yielding high PSNR values.,,We propose a novel application of automated texture synthesis in combination with a perceptual loss focusing on creating realistic textures rather than optimizing for a pixelaccurate reproduction of ground truth images during training. By using feed-forward fully convolutional neural networks in an adversarial training setting, we achieve a significant boost in image quality at high magnification ratios. Extensive experiments on a number of datasets show the effectiveness of our approach, yielding state-of-the-art results in both quantitative and qualitative benchmarks.
ISSN:2380-7504
DOI:10.1109/ICCV.2017.481