Accurate Image Super-Resolution Using Very Deep Convolutional Networks

TL;DR

This paper introduces a very deep convolutional network for single-image super-resolution, significantly improving accuracy by leveraging residual learning, high learning rates, and a novel training procedure.

Contribution

The paper proposes a 20-layer deep CNN for super-resolution that exploits residual learning and high learning rates with gradient clipping to enhance accuracy and training efficiency.

Findings

Outperforms existing super-resolution methods in accuracy

Achieves noticeable visual improvements in reconstructed images

Utilizes very deep networks with effective training strategies

Abstract

We present a highly accurate single-image super-resolution (SR) method. Our method uses a very deep convolutional network inspired by VGG-net used for ImageNet classification \cite{simonyan2015very}. We find increasing our network depth shows a significant improvement in accuracy. Our final model uses 20 weight layers. By cascading small filters many times in a deep network structure, contextual information over large image regions is exploited in an efficient way. With very deep networks, however, convergence speed becomes a critical issue during training. We propose a simple yet effective training procedure. We learn residuals only and use extremely high learning rates ($10^4$ times higher than SRCNN \cite{dong2015image}) enabled by adjustable gradient clipping. Our proposed method performs better than existing methods in accuracy and visual improvements in our results are easily noticeable.