Real Image Super-Resolution using GAN through modeling of LR and HR process

TL;DR

This paper introduces a GAN-based super-resolution method that models real-world image degradations by learning degradation processes, enabling more accurate super-resolution of real images compared to traditional methods.

Contribution

It proposes a novel GAN framework with learnable nonlinearities to model complex real-world degradations for training super-resolution models.

Findings

Effective in handling real-world image degradations

Outperforms traditional bicubic-based SR methods

Produces high-quality super-resolved images

Abstract

The current existing deep image super-resolution methods usually assume that a Low Resolution (LR) image is bicubicly downscaled of a High Resolution (HR) image. However, such an ideal bicubic downsampling process is different from the real LR degradations, which usually come from complicated combinations of different degradation processes, such as camera blur, sensor noise, sharpening artifacts, JPEG compression, and further image editing, and several times image transmission over the internet and unpredictable noises. It leads to the highly ill-posed nature of the inverse upscaling problem. To address these issues, we propose a GAN-based SR approach with learnable adaptive sinusoidal nonlinearities incorporated in LR and SR models by directly learn degradation distributions and then synthesize paired LR/HR training data to train the generalized SR model to real image degradations. We demonstrate the effectiveness of our proposed approach in quantitative and qualitative experiments.