A comprehensive review of image super-resolution metrics: classical and AI-based approaches

Abstract

Image super-resolution is a process that aims to enhance the quality and resolution of images using various techniques and algorithms. The process aims to reconstruct a high-resolution image from a given low-resolution input. To determine the effectiveness of these algorithms, it's crucial to evaluate those using specific metrics. In this paper, we take a closer look at the most commonly used image super-resolution metrics, including classical approaches like Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Peak Signal to Noise Ratio (PSNR), and Structural Similarity Index (SSIM). We also discuss advanced metrics like Learned Perceptual Image Patch Similarity (LPIPS), Fréchet Inception Distance (FID), Inception Score (IS), and Multi-Scale Structural Similarity Index (MS-SSIM). Furthermore, we provide an overview of classical and AI-based super-resolution techniques and methods. Finally, we discuss potential challenges and future research directions in the field and present our experimental results by applying image super-resolution metrics. In the result and discussion section, we have practiced some given metrics and proposed our image super-resolution results.