Image Super-Resolution Using Deep Convolutional Networks

TLDR

Traditional sparse‑coding based SR methods can be interpreted as deep convolutional networks. The study proposes a deep learning approach for single‑image super‑resolution. The authors design a deep CNN that learns an end‑to‑end mapping from low‑ to high‑resolution images, jointly optimizes all layers, explores various network structures and parameter settings, and extends the model to handle three color channels simultaneously. The lightweight CNN delivers state‑of‑the‑art restoration quality and fast online speed, and the multi‑channel extension further improves overall reconstruction quality.

Abstract

We propose a deep learning method for single image super-resolution (SR). Our method directly learns an end-to-end mapping between the low/high-resolution images. The mapping is represented as a deep convolutional neural network (CNN) that takes the low-resolution image as the input and outputs the high-resolution one. We further show that traditional sparse-coding-based SR methods can also be viewed as a deep convolutional network. But unlike traditional methods that handle each component separately, our method jointly optimizes all layers. Our deep CNN has a lightweight structure, yet demonstrates state-of-the-art restoration quality, and achieves fast speed for practical on-line usage. We explore different network structures and parameter settings to achieve trade-offs between performance and speed. Moreover, we extend our network to cope with three color channels simultaneously, and show better overall reconstruction quality.