Cross-modality interactive attention network for multispectral pedestrian detection

doi:10.1016/j.inffus.2018.09.015

	Cross-modality interactive attention network for multispectral pedestrian detection
	Zhang, Lu 1,4; Liu, Zhiyong1,4,7 ; Zhang, Shifeng2,3,4 ; Yang, Xu1,4 ; Qiao, Hong1,4,7 ; Huang, Kaizhu 5; Hussain, Amir 6
刊名	INFORMATION FUSION
	2019-10-01
卷号	50 页码:20-29
关键词	Pedestrian detection Modality fusion Cross-modality attention Deep neural networks
ISSN号	1566-2535
DOI	10.1016/j.inffus.2018.09.015
通讯作者	Liu, Zhiyong(zhiyong.liu@ia.ac.cn)
英文摘要	Multispectral pedestrian detection is an emerging solution with great promise in many around-the-clock applications, such as automotive driving and security surveillance. To exploit the complementary nature and remedy contradictory appearance between modalities, in this paper, we propose a novel cross-modality interactive attention network that takes full advantage of the interactive properties of multispectral input sources. Specifically, we first utilize the color (RGB) and thermal streams to build up two detached feature hierarchy for each modality, then by taking the global features, correlations between two modalities are encoded in the attention module. Next, the channel responses of halfway feature maps are recalibrated adaptively for subsequent fusion operation. Our architecture is constructed in the multi-scale format to better deal with different scales of pedestrians, and the whole network is trained in an end-to-end way. The proposed method is extensively evaluated on the challenging KAIST multispectral pedestrian dataset and achieves state-of-the-art performance with high efficiency.
资助项目	National Key Research and Development Plan of China[2017YFB1300202] ; National Key Research and Development Plan of China[2016YFC0300801] ; National Natural Science Foundation of China[U1613213] ; National Natural Science Foundation of China[61627808] ; National Natural Science Foundation of China[61503383] ; National Natural Science Foundation of China[61210009] ; National Natural Science Foundation of China[91648205] ; National Natural Science Foundation of China[61702516] ; National Natural Science Foundation of China[61473236] ; National Natural Science Foundation of China[61876155] ; Ministry of Science and Technology of the People's Republic of China[2015BAK35B00] ; Ministry of Science and Technology of the People's Republic of China[2015BAK35B01] ; Chinese Academy of Sciences (Science Frontier Program)[XDBS01050100] ; Natural Science Foundation of the Jiangsu Higher Education Institutions of China[17KJD520010] ; Guangdong Science and Technology Department[2016B090910001] ; Suzhou Science and Technology Program[SYG201712] ; Suzhou Science and Technology Program[SZS201613] ; Key Program Special Fund in XJTLU[KSF-A-01] ; Key Program Special Fund in XJTLU[KSF-P-02] ; UK Engineering and Physical Sciences Research Council (EPSRC)[EP/M026981/1]
WOS关键词	FUSION ; IMAGES
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER SCIENCE BV
WOS记录号	WOS:000466056900003
资助机构	National Key Research and Development Plan of China ; National Natural Science Foundation of China ; Ministry of Science and Technology of the People's Republic of China ; Chinese Academy of Sciences (Science Frontier Program) ; Natural Science Foundation of the Jiangsu Higher Education Institutions of China ; Guangdong Science and Technology Department ; Suzhou Science and Technology Program ; Key Program Special Fund in XJTLU ; UK Engineering and Physical Sciences Research Council (EPSRC)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/24394]
专题	中国科学院自动化研究所
通讯作者	Liu, Zhiyong
作者单位	1.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automat, Ctr Biometr & Secur Res, Beijing, Peoples R China 3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China 4.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 5.Xian Jiaotong Liverpool Univ, Dept EEE, SIP, Renai Rd 111, Suzhou 215123, Jiangsu, Peoples R China 6.Edinburgh Napier Univ, Sch Comp, Merchiston Campus, Edinburgh EH10 5DT, Midlothian, Scotland 7.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Lu,Liu, Zhiyong,Zhang, Shifeng,et al. Cross-modality interactive attention network for multispectral pedestrian detection[J]. INFORMATION FUSION,2019,50:20-29.
APA	Zhang, Lu.,Liu, Zhiyong.,Zhang, Shifeng.,Yang, Xu.,Qiao, Hong.,...&Hussain, Amir.(2019).Cross-modality interactive attention network for multispectral pedestrian detection.INFORMATION FUSION,50,20-29.
MLA	Zhang, Lu,et al."Cross-modality interactive attention network for multispectral pedestrian detection".INFORMATION FUSION 50(2019):20-29.