Large Receptive Field Networks for High-Scale Image Super-Resolution

TL;DR

This paper introduces Large Receptive Field Networks for high-scale image super-resolution, expanding receptive fields without increasing model size, leading to improved performance and efficiency in super-resolution tasks.

Contribution

The paper proposes novel methods using 1-D separable kernels and atrous convolutions to enlarge receptive fields without adding parameters or depth, enhancing super-resolution performance.

Findings

Effective at high upscaling factors

Improved PSNR and SSIM scores

Reduced model complexity and training difficulty

Abstract

Convolutional Neural Networks have been the backbone of recent rapid progress in Single-Image Super-Resolution. However, existing networks are very deep with many network parameters, thus having a large memory footprint and being challenging to train. We propose Large Receptive Field Networks which strive to directly expand the receptive field of Super-Resolution networks without increasing depth or parameter count. In particular, we use two different methods to expand the network receptive field: 1-D separable kernels and atrous convolutions. We conduct considerable experiments to study the performance of various arrangement schemes of the 1-D separable kernels and atrous convolution in terms of accuracy (PSNR / SSIM), parameter count, and speed, while focusing on the more challenging high upscaling factors. Extensive benchmark evaluations demonstrate the effectiveness of our approach.