CASIA OpenIR  > 模式识别国家重点实验室  > 生物识别与安全技术
An Efficient Multiresolution Network for Vehicle Reidentification
Shen, Fei1; Zhu, Jianqing1; Zhu, Xiaobin2; Huang, Jingchang3; Zeng, Huanqiang1; Lei, Zhen4,5,6; Cai, Canhui1
Source PublicationIEEE INTERNET OF THINGS JOURNAL
ISSN2327-4662
2022-06-01
Volume9Issue:11Pages:9049-9059
Corresponding AuthorZhu, Jianqing(jqzhu@hqu.edu.cn) ; Zeng, Huanqiang(zeng0043@hqu.edu.cn)
AbstractIn general, vehicle images have varying resolutions due to vehicles' movements and different camera settings. However, most existing vehicle reidentification models are single-resolution deep networks trained with preuniformly resizing vehicle images, which underestimate adverse effects of varying resolutions and lead to unsatisfactory performance. A straightforward solution for dealing with varying resolutions is to train multiple vehicle reidentification models. Each model is independently trained with images of a specific resolution. However, this straightforward solution requires significant overhead and ignores intrinsic associations among different resolution images. For that, an efficient multiresolution network (EMRN) is proposed for vehicle reidentification in this article. First, EMRN embeds a newly designed multiresolution feature dimension uniform module (MR-FDUM) behind a traditional backbone network (i.e., ResNet-50). As a result, the whole model can extract fixed dimensional features from different resolution images so that it can be trained with one loss function of fixed dimensional parameters rather than training multiple models. Second, a multiresolution image randomly feeding strategy is designed to train EMRN, making each minibatch data of a random resolution during the training process. Consequently, EMRN can implicitly learn collaborative multiresolution features via only a unitary deep network. The experiments on three large-scale data sets, i.e., VeRi776, VehicleID, and VRIC, demonstrate that EMRN is superior to state-of-the-art vehicle reidentification methods.
KeywordImage resolution Training Feature extraction Spatial resolution Proposals Internet of Things Deep learning Deep learning image representation multiresolution vehicle reidentification
DOI10.1109/JIOT.2021.3119525
Indexed BySCI
Language英语
Funding ProjectNational Key Research and Development Program[2020YFC2003901] ; National Natural Science Foundation of China[61976098] ; National Natural Science Foundation of China[61871434] ; National Natural Science Foundation of China[61802136] ; National Natural Science Foundation of China[6217070593] ; Natural Science Foundation for Outstanding Young Scholars of Fujian Province[2019J06017] ; Key Science and Technology Project of Xiamen City[3502ZCQ20191005] ; Science and Technology Bureau of Quanzhou[2018C115R] ; Postgraduates' Innovative Fund in Scientific Research of Huaqiao University[18014084008]
Funding OrganizationNational Key Research and Development Program ; National Natural Science Foundation of China ; Natural Science Foundation for Outstanding Young Scholars of Fujian Province ; Key Science and Technology Project of Xiamen City ; Science and Technology Bureau of Quanzhou ; Postgraduates' Innovative Fund in Scientific Research of Huaqiao University
WOS Research AreaComputer Science ; Engineering ; Telecommunications
WOS SubjectComputer Science, Information Systems ; Engineering, Electrical & Electronic ; Telecommunications
WOS IDWOS:000800215600092
PublisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/49567
Collection模式识别国家重点实验室_生物识别与安全技术
Corresponding AuthorZhu, Jianqing; Zeng, Huanqiang
Affiliation1.Huaqiao Univ, Coll Engn, Quanzhou 362021, Peoples R China
2.Univ Sci & Technol Beijing, Dept Comp Sci & Technol, Beijing 100083, Peoples R China
3.Chinese Acad Sci, Shanghai Inst Microsyst & Informat Technol, Shanghai 200050, Peoples R China
4.Chinese Acad Sci, Ctr Biometr & Secur Res, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
5.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
6.Chinese Acad Sci, Ctr Artificial Intelligence & Robot, Hong Kong Inst Sci & Innovat, Hong Kong, Peoples R China
Recommended Citation
GB/T 7714
Shen, Fei,Zhu, Jianqing,Zhu, Xiaobin,et al. An Efficient Multiresolution Network for Vehicle Reidentification[J]. IEEE INTERNET OF THINGS JOURNAL,2022,9(11):9049-9059.
APA Shen, Fei.,Zhu, Jianqing.,Zhu, Xiaobin.,Huang, Jingchang.,Zeng, Huanqiang.,...&Cai, Canhui.(2022).An Efficient Multiresolution Network for Vehicle Reidentification.IEEE INTERNET OF THINGS JOURNAL,9(11),9049-9059.
MLA Shen, Fei,et al."An Efficient Multiresolution Network for Vehicle Reidentification".IEEE INTERNET OF THINGS JOURNAL 9.11(2022):9049-9059.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Shen, Fei]'s Articles
[Zhu, Jianqing]'s Articles
[Zhu, Xiaobin]'s Articles
Baidu academic
Similar articles in Baidu academic
[Shen, Fei]'s Articles
[Zhu, Jianqing]'s Articles
[Zhu, Xiaobin]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Shen, Fei]'s Articles
[Zhu, Jianqing]'s Articles
[Zhu, Xiaobin]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.