Dually Connected Deraining Net Using Pixel-Wise Attention | |
Ren WH(任卫红)1,2![]() ![]() ![]() | |
Department | 机器人学研究室 |
Source Publication | IEEE Signal Processing Letters
![]() |
ISSN | 1070-9908 |
2020 | |
Volume | 27Pages:316-320 |
Indexed By | SCI ; EI |
EI Accession number | 20201008257947 |
WOS ID | WOS:000526664500007 |
Contribution Rank | 1 |
Funding Organization | National Natural Science Foundation of China under Grant 91648118, Grant 61991413, and Grant 61821005 ; LiaoNing Revitalization Talents Program |
Keyword | Image deraining pixel-wise attention encoderdecoder skip connection |
Abstract | Recent single image deraining methods either use a recurrent mechanism to gradually learn the mapping between clear images and rainy images, or focus on designing various loss functions to supervise the learning process. In this letter, we propose a dually connected deraining net using pixel-wise attention, for single image rain removal. Specifically, the deraining net adopts an encoder-decoder net as a backbone, which can effectively learn a residual rain-streaks map by jointly using skip sum connection and skip concatenation connection. The dual connections enable the deraining net to promote information flow between layers, and thus can allow it to discriminate and localize the rain streaks. To preserve image details, the decoded features are weighted by the learnable pixel-wise attention for adaptively recalibrating their responses. Experimental results on synthetic datasets demonstrate that the proposed model outperforms the recent state-of-the-art deraining methods. |
Language | 英语 |
WOS Subject | Engineering, Electrical & Electronic |
WOS Research Area | Engineering |
Funding Project | National Natural Science Foundation of China[91648118] ; National Natural Science Foundation of China[61991413] ; National Natural Science Foundation of China[61821005] ; LiaoNing Revitalization Talents Program |
Citation statistics | |
Document Type | 期刊论文 |
Identifier | http://ir.sia.cn/handle/173321/26441 |
Collection | 机器人学研究室 |
Corresponding Author | Tian JD(田建东) |
Affiliation | 1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China 2.University of Chinese Academy of Sciences, Huairou 100049, China |
Recommended Citation GB/T 7714 | Ren WH,Tian JD,Wang Q,et al. Dually Connected Deraining Net Using Pixel-Wise Attention[J]. IEEE Signal Processing Letters,2020,27:316-320. |
APA | Ren WH,Tian JD,Wang Q,&Tang YD.(2020).Dually Connected Deraining Net Using Pixel-Wise Attention.IEEE Signal Processing Letters,27,316-320. |
MLA | Ren WH,et al."Dually Connected Deraining Net Using Pixel-Wise Attention".IEEE Signal Processing Letters 27(2020):316-320. |
Files in This Item: | ||||||
File Name/Size | DocType | Version | Access | License | ||
Dually Connected Der(2032KB) | 期刊论文 | 出版稿 | 开放获取 | CC BY-NC-SA | View Application Full Text |
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment