中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 地理科学与资源研究所
  3. 中国科学院地理科学与资源研究所
Internal Learning for Image Super-Resolution by Adaptive Feature Transform

文献类型：期刊论文

作者He, Yifan1; Cao, Wei1; Du, Xiaofeng1; Chen, Changlin2
刊名SYMMETRY-BASEL
出版日期2020-10-01
卷号12期号:10页码:19
关键词super-resolution internal learning feature transform deep convolutional neural network
DOI10.3390/sym12101686
通讯作者Du, Xiaofeng(xfdu@xmut.edu.cn)
英文摘要Recent years have witnessed the great success of image super-resolution based on deep learning. However, it is hard to adapt a well-trained deep model for a specific image for further improvement. Since the internal repetition of patterns is widely observed in visual entities, internal self-similarity is expected to help improve image super-resolution. In this paper, we focus on exploiting a complementary relation between external and internal example-based super-resolution methods. Specifically, we first develop a basic network learning external prior from large scale training data and then learn the internal prior from the given low-resolution image for task adaptation. By simply embedding a few additional layers into a pre-trained deep neural network, the image-adaptive super-resolution method exploits the internal prior for a specific image, and the external prior from a well-trained super-resolution model. We achieve 0.18 dB PSNR improvements over the basic network's results on standard datasets. Extensive experiments under image super-resolution tasks demonstrate that the proposed method is flexible and can be integrated with lightweight networks. The proposed method boosts the performance for images with repetitive structures, and it improves the accuracy of the reconstructed image of the lightweight model.
资助项目National Natural Science Foundation of China[61806173] ; Natural Science Foundation of Fujian Province of China[2019J01855] ; Natural Science Foundation of Fujian Province of China[2019J01854] ; Scientific Research Foundation of Xiamen for the Returned Overseas Chinese Scholars (XRS[2018])[310] ; Scientific Research Fund of Fujian Provincial Education Department, China[JT180440] ; Science and Technology Program of Xiamen, China[3502Z20179032]
WOS研究方向Science & Technology - Other Topics
语种英语
WOS记录号WOS:000585201400001
出版者MDPI
资助机构National Natural Science Foundation of China ; Natural Science Foundation of Fujian Province of China ; Scientific Research Foundation of Xiamen for the Returned Overseas Chinese Scholars (XRS[2018]) ; Scientific Research Fund of Fujian Provincial Education Department, China ; Science and Technology Program of Xiamen, China
源URL[http://ir.igsnrr.ac.cn/handle/311030/156577]  
专题中国科学院地理科学与资源研究所
通讯作者Du, Xiaofeng
作者单位1.Xiamen Univ Technol, Sch Comp & Informat Engn, Xiamen 361024, Peoples R China
2.Chinese Acad Sci, Inst Geog Sci & Nat Resources, State Key Lab Resources & Environm Informat Syst, Beijing 100864, Peoples R China
推荐引用方式
GB/T 7714		 He, Yifan,Cao, Wei,Du, Xiaofeng,et al. Internal Learning for Image Super-Resolution by Adaptive Feature Transform[J]. SYMMETRY-BASEL,2020,12(10):19.
APA He, Yifan,Cao, Wei,Du, Xiaofeng,&Chen, Changlin.(2020).Internal Learning for Image Super-Resolution by Adaptive Feature Transform.SYMMETRY-BASEL,12(10),19.
MLA He, Yifan,et al."Internal Learning for Image Super-Resolution by Adaptive Feature Transform".SYMMETRY-BASEL 12.10(2020):19.

入库方式： OAI收割

来源：地理科学与资源研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。