Application of convolutional neural networks in image super-resolution

TL;DR

This paper reviews various CNN-based methods for image super-resolution, analyzing their principles, differences, and performance through experiments, to facilitate future developments in the field.

Contribution

It provides a comprehensive summary and comparison of CNN-based super-resolution techniques, highlighting their principles, differences, and experimental performance.

Findings

Different CNN-based interpolation methods have varying performance and computational efficiency.

Analysis of CNN modules reveals their impact on super-resolution quality.

The paper identifies potential research directions and current limitations.

Abstract

Due to strong learning abilities of convolutional neural networks (CNNs), they have become mainstream methods for image super-resolution. However, there are big differences of different deep learning methods with different types. There is little literature to summarize relations and differences of different methods in image super-resolution. Thus, summarizing these literatures are important, according to loading capacity and execution speed of devices. This paper first introduces principles of CNNs in image super-resolution, then introduces CNNs based bicubic interpolation, nearest neighbor interpolation, bilinear interpolation, transposed convolution, sub-pixel layer, meta up-sampling for image super-resolution to analyze differences and relations of different CNNs based interpolations and modules, and compare performance of these methods by experiments. Finally, this paper gives potential research points and drawbacks and summarizes the whole paper, which can facilitate developments of CNNs in image super-resolution.