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.

Keywords

Computer scienceConvolutional neural networkArtificial intelligenceDeep learningPattern recognition (psychology)Image (mathematics)Coding (social sciences)SuperresolutionImage resolutionComputer visionMathematics

Affiliated Institutions

Related Publications

Publication Info

Year
2015
Type
article
Volume
38
Issue
2
Pages
295-307
Citations
9271
Access
Closed

External Links

Social Impact

Social media, news, blog, policy document mentions

Citation Metrics

9271
OpenAlex

Cite This

Chao Dong, Chen Change Loy, Kaiming He et al. (2015). Image Super-Resolution Using Deep Convolutional Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence , 38 (2) , 295-307. https://doi.org/10.1109/tpami.2015.2439281

Identifiers

DOI
10.1109/tpami.2015.2439281