Large Receptive Field Networks for High-Scale Image Super-Resolution
This addresses the challenge of memory footprint and training difficulty in super-resolution networks, though it appears incremental by focusing on architectural modifications.
The paper tackles the problem of high-scale image super-resolution by proposing Large Receptive Field Networks to expand receptive fields without increasing depth or parameters, achieving effectiveness in benchmark evaluations for high upscaling factors.
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.