{
  "id": "1501.00092",
  "version": "v1",
  "published": "2014-12-31T08:35:09.000Z",
  "updated": "2014-12-31T08:35:09.000Z",
  "title": "Image Super-Resolution Using Deep Convolutional Networks",
  "authors": [
    "Chao Dong",
    "Chen Change Loy",
    "Kaiming He",
    "Xiaoou Tang"
  ],
  "comment": "14 pages, 14 figures, journal",
  "categories": [
    "cs.CV",
    "cs.NE"
  ],
  "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.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2014-12-31T08:35:09.000Z"
    }
  ],
  "analyses": {
    "subjects": [
      "I.4.5",
      "I.2.6"
    ],
    "keywords": [
      "deep convolutional network",
      "deep convolutional neural network",
      "better overall reconstruction quality",
      "demonstrates state-of-the-art restoration quality",
      "traditional sparse-coding-based sr methods"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 14,
      "language": "en",
      "license": "arXiv",
      "status": "editable",
      "adsabs": "2015arXiv150100092D"
    }
  }
}