SMNet: Synchronous Multi-Scale Low Light Enhancement Network With Local and Global Concern

doi:10.1109/TMM.2023.3254141

CSpace

	SMNet: Synchronous Multi-Scale Low Light Enhancement Network With Local and Global Concern
	Lin, Shideng 1,2; Tang, Fan 3; Dong, Weiming 1,2; Pan, Xingjia 4; Xu, Changsheng 1,2
	2023
发表期刊	IEEE TRANSACTIONS ON MULTIMEDIA
ISSN	1520-9210
卷号	25 页码:9506-9517
摘要	Limited by objectively poor lighting conditions and hardware devices, low-light images with low visual quality and low visibility are inevitable in the real world. Accurate local details and reasonable global information play their essential and distinct roles in low-light image enhancement: local details contribute to fine textures, while global information is critical for a proper understanding of the global brightness level. In this article, we focus on integrating local and global aspects to achieve high-quality low-light image enhancement by proposing the synchronous multi-scale low-light enhancement network (SMNet). A synchronous multi-scale representation learning structure and a global feature recalibration module are adopted in SMNet. Different from the traditional multi-scale feature learning architecture, SMNet carries out the multi-scale representation learning in a synchronous way: we first calculate the rough contextual representations in a top-down manner and then learn multi-scale representations in a bottom-up way to generate representations with rich local details. To acquire global brightness information, a global feature recalibration module (GFRM) is applied after the synchronous multi-scale representations to perceive and exploit proper global information by global pooling and projection to recalibrate channel weights globally. The synchronous multi-scale representation and GFRM compose the basic local-and-global block. Experimental results on mainstream real-world dataset LOL and synthetic dataset MIT-Adobe FiveK show that the proposed SMNet not only leads the way on objective metrics (0.41/2.31 improvement of PSNR on two datasets) but is also superior in subjective comparisons compared with typical SoTA methods.
关键词	Low-light image enhancement multi-scale feature learning deep-learning
DOI	10.1109/TMM.2023.3254141
收录类别	SCI
语种	英语
资助项目	Beijing Natural Science Foundation
WOS研究方向	Computer Science ; Telecommunications
WOS类目	Computer Science, Information Systems ; Computer Science, Software Engineering ; Telecommunications
WOS记录号	WOS:001133324200019
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/38430
专题	中国科学院计算技术研究所
通讯作者	Tang, Fan
作者单位	1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 101408, Peoples R China 2.Chinese Acad Sci, Inst Automat, NLPR, Beijing 101408, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 4.Tencent, Youtu Lab, Shanghai 200001, Peoples R China
推荐引用方式 GB/T 7714	Lin, Shideng,Tang, Fan,Dong, Weiming,et al. SMNet: Synchronous Multi-Scale Low Light Enhancement Network With Local and Global Concern[J]. IEEE TRANSACTIONS ON MULTIMEDIA,2023,25:9506-9517.
APA	Lin, Shideng,Tang, Fan,Dong, Weiming,Pan, Xingjia,&Xu, Changsheng.(2023).SMNet: Synchronous Multi-Scale Low Light Enhancement Network With Local and Global Concern.IEEE TRANSACTIONS ON MULTIMEDIA,25,9506-9517.
MLA	Lin, Shideng,et al."SMNet: Synchronous Multi-Scale Low Light Enhancement Network With Local and Global Concern".IEEE TRANSACTIONS ON MULTIMEDIA 25(2023):9506-9517.