Single Image Super-Resolution via Multi-Scale Information Polymerization Network

Recently, the performances of deep convolution neural networks (CNNs)-based single-image super-resolution (SISR) have been significantly improved. However, most of the existing CNN-based SISR methods mainly focus on wider or deeper networks and ignore the potential relationship between multi-scale features, leading to the limited representation ability of the reconstructed network. To address this problem, we propose a new multi-scale information polymerization network (MIPN). Specifically, we propose a multi-scale information polymerization block (MIPB), which uses convolution layers of different convolution kernel sizes to extract multi-scale image features, and effectively polymerizate the extracted features together to obtain fine image features. Moreover, we also propose a shallow residual block in MIPB. Compared with the traditional convolution layer, this proposed block can effectively extract image features without increasing the number of parameters. Extensive experiments show that the proposed method performs better than several state-of-the-art methods in quantitative and visual quality indicators.

I. Introduction

Single-image super-resolution (SISR) is a classic computer vision task, which aims to reconstruct the corresponding high-resolution (HR) image from a given low-resolution (LR) image. As a low-level problem of image processing, super-resolution (SR) has been paid attention by more and more researchers. And SR methods have been widely used in other advanced visual tasks, such as video codecs [1], security and surveillance imaging [2], satellite and aerial imaging [3], and facial analysis [4]. Therefore, how to reconstruct high-quality SR images is of great significance.