Deep Residual Networks with a Fully Connected Recon-struction Layer for Single Image Super-Resolution

TL;DR

This paper proposes a novel single image super-resolution architecture that uses a fully connected reconstruction layer to better exploit global contextual information while maintaining low computational complexity, and introduces an edge difference constraint to enhance edge and texture preservation.

Contribution

The paper introduces a new SISR model with a fully connected layer for differentiated upsampling and an edge difference loss to improve reconstruction quality.

Findings

Outperforms state-of-the-art methods in accuracy

Effectively exploits global contextual information

Maintains low computational complexity

Abstract

Recently, deep neural networks have achieved impressive performance in terms of both reconstruction accuracy and efficiency for single image super-resolution (SISR). However, the network model of these methods is a fully convolutional neural network, which is limit to exploit the differentiated contextual information over the global region of the input image because of the weight sharing in convolution height and width extent. In this paper, we discuss a new SISR architecture where features are extracted in the low-resolution (LR) space, and then we use a fully connected layer which learns an array of differentiated upsampling weights to reconstruct the desired high-resolution (HR) image from the final obtained LR features. By doing so, we effectively exploit the differentiated contextual information over the whole input image region, whilst maintaining the low computational complexity for the overall SR operations. In addition, we introduce an edge difference constraint into our loss function to preserve edges and texture structures. Extensive experiments validate that our SISR method outperforms the existing state-of-the-art methods.