Adaptive super -resolution for person re-identification with low-resolution images

doi:10.1016/j.patcog.2020.107682

CORC > 自动化研究所 > 中国科学院自动化研究所 > 智能感知与计算研究中心

	Adaptive super -resolution for person re-identification with low-resolution images
	Han, Ke 2,3; Huang, Yan 2; Song, Chunfeng 2; Wang, Liang 1,2,4; Tan, Tieniu 1,2
刊名	PATTERN RECOGNITION
	2021-06-01
卷号	114 页码:12
关键词	Person re-identification Super-resolution Body regions Adaptive feature integration
ISSN号	0031-3203
DOI	10.1016/j.patcog.2020.107682
通讯作者	Huang, Yan(yhuang@nlpr.ia.ac.cn)
英文摘要	Person re-identification is challenging with low-resolution query and high-resolution gallery images. To address the resolution mismatch, many methods perform super-resolution (SR) on low-resolution queries with specifying a single scale factor. However, using a single SR module, whichever scale factor is speci-fied, always brings both advantages and drawbacks in recovering and identifying identity information. A larger scale factor recovers more details but produces excessive artifacts, while a smaller one is on the contrary. To exploit their complementary property for more robust recovery and identification, we pro-pose the Adaptive Person Super-Resolution (APSR) model. APSR jointly trains and fuses multiple SR mod-ules based on their generated visual contents, to fully compensate and learn the complementary identity features in an end-to-end manner. To improve the robustness to artifacts during fusion, our model fur-ther learns informative features by online dividing and integrating the generated body regions. Extensive experiments verify the effectiveness of our method with state-of-the-art performances. ? 2021 Elsevier Ltd. All rights reserved.
资助项目	National Key Research and Development Program of China[2016YFB1001000] ; Key Research Program of Frontier Sciences, CAS[ZDBS-LY-JSC032] ; Shandong Provincial Key Research and Development Program[2019JZZY010119] ; CAS-AIR
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	ELSEVIER SCI LTD
WOS记录号	WOS:000632383600004
资助机构	National Key Research and Development Program of China ; Key Research Program of Frontier Sciences, CAS ; Shandong Provincial Key Research and Development Program ; CAS-AIR
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/44148]
专题	自动化研究所_智能感知与计算研究中心
通讯作者	Huang, Yan
作者单位	1.Ctr Excellence Brain Sci & Intelligence Technol C, Beijing, Peoples R China 2.Univ Chinese Acad Sci UCAS, Ctr Res Intelligent Percept & Comp CRIPAC, Natl Lab Pattern Recognit NLPR, Inst Automat,Chinese Acad Sci CASIA, Beijing, Peoples R China 3.Univ Chinese Acad Sci UCAS, Sch Future Technol, Beijing, Peoples R China 4.Chinese Acad Sci, Artificial Intelligence Res CAS AIR, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Han, Ke,Huang, Yan,Song, Chunfeng,et al. Adaptive super -resolution for person re-identification with low-resolution images[J]. PATTERN RECOGNITION,2021,114:12.
APA	Han, Ke,Huang, Yan,Song, Chunfeng,Wang, Liang,&Tan, Tieniu.(2021).Adaptive super -resolution for person re-identification with low-resolution images.PATTERN RECOGNITION,114,12.
MLA	Han, Ke,et al."Adaptive super -resolution for person re-identification with low-resolution images".PATTERN RECOGNITION 114(2021):12.