Seeing Through Darkness: Visual Localization at Night via Weakly Supervised Learning of Domain Invariant Features

doi:10.1109/TMM.2022.3154165

CASIA OpenIR > 离退休人员

	Seeing Through Darkness: Visual Localization at Night via Weakly Supervised Learning of Domain Invariant Features
	Fan, Bin1 ; Yang, Yuzhu 1; Feng, Wensen 2; Wu, Fuchao3 ; Lu, Jiwen 4; Liu, Hongmin 1
发表期刊	IEEE TRANSACTIONS ON MULTIMEDIA
ISSN	1520-9210
	2023
卷号	25 页码:1713-1726
通讯作者	Liu, Hongmin(hmliu_82@163.com)
摘要	Long term visual localization has to conquer the problem of matching images with dramatic photometric changes caused by different seasons, natural and man-made illumination changes, etc. Visual localization at night plays a vital role in many applications like autonomous driving and augmented reality, for which extracting keypoints and descriptors with robustness to day-night illumination changes has became the bottleneck. This paper proposes an adversarial learning based solution to harvest from the weakly domain labels of day and night images, along with the point level correspondences among day time images, to achieve robust local feature extraction and description across day-night images. The key idea is to learn a discriminator to distinguish whether a feature map is generated from the day or night images, and simultaneously to adjust the parameters of feature extraction network so as to fool the discriminator. After adversarial training of the discriminator and feature extraction network, the feature extraction network finally reaches a stable status so that the extracted feature maps are robust to day-night photometric changes, based on which day-night domain invariant keypoints and descriptors can be extracted. Compared to existing local feature learning methods, it only requires an additional set of easily captured night images to improve the domain invariance of learned features. Experiments on two challenging benchmarks show the effectiveness of proposed method. In addition, this paper revisits the widely used image matching metrics on HPatches and finds that recall of different methods is highly related to their relative localization performance.
关键词	Domain invariant local features image matching long-term visual localization weakly supervised learning
DOI	10.1109/TMM.2022.3154165
关键词[WOS]	OBJECT DETECTION ; ACCURATE ; DESCRIPTORS ; SLAM
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2020YFB1313002] ; National Natural Science Foundation of China[61876180] ; National Natural Science Foundation of China[U2013202] ; National Natural Science Foundation of China[62076026] ; Beijing Natural Science Foundation[4202073] ; Guangdong Basic and Applied Basic Research Foundation[2020B1515120050] ; Fundamental Research Funds for the Central Universities[FRF-TP-20-08B] ; CAAI-Huawei MindSpore Open Fund
项目资助者	National Key Research and Development Program of China ; National Natural Science Foundation of China ; Beijing Natural Science Foundation ; Guangdong Basic and Applied Basic Research Foundation ; Fundamental Research Funds for the Central Universities ; CAAI-Huawei MindSpore Open Fund
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:001007432100009
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：17[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/53703
专题	离退休人员
通讯作者	Liu, Hongmin
作者单位	1.Univ Sci & Technol Beijing, Sch Automat & Elect Engn, Beijing 100083, Peoples R China 2.Cent Media Technol Inst, Shenzhen 518129, Peoples R China 3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 4.Tsinghua Univ, Dept Automat, Beijing 100084, Peoples R China
推荐引用方式 GB/T 7714	Fan, Bin,Yang, Yuzhu,Feng, Wensen,et al. Seeing Through Darkness: Visual Localization at Night via Weakly Supervised Learning of Domain Invariant Features[J]. IEEE TRANSACTIONS ON MULTIMEDIA,2023,25:1713-1726.
APA	Fan, Bin,Yang, Yuzhu,Feng, Wensen,Wu, Fuchao,Lu, Jiwen,&Liu, Hongmin.(2023).Seeing Through Darkness: Visual Localization at Night via Weakly Supervised Learning of Domain Invariant Features.IEEE TRANSACTIONS ON MULTIMEDIA,25,1713-1726.
MLA	Fan, Bin,et al."Seeing Through Darkness: Visual Localization at Night via Weakly Supervised Learning of Domain Invariant Features".IEEE TRANSACTIONS ON MULTIMEDIA 25(2023):1713-1726.