Unsupervised Nighttime Object Tracking Based on Transformer and Domain Adaptation Fusion Network
In view of the complexity, uncertainty, and low visibility of the night-time environment. In this study, we designed an unsupervised domain adaptation framework, TransffCAR, using a Siamese network of the source (daytime) and target (nighttime) domains. Training patches are generated by finding targ...
Saved in:
Published in | IEEE access Vol. 12; pp. 130896 - 130913 |
---|---|
Main Authors | , , , , , |
Format | Journal Article |
Language | English |
Published |
IEEE
2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | In view of the complexity, uncertainty, and low visibility of the night-time environment. In this study, we designed an unsupervised domain adaptation framework, TransffCAR, using a Siamese network of the source (daytime) and target (nighttime) domains. Training patches are generated by finding targets from a large number of unlabelled frame images. This avoids heavy manual labelling, improves the performance of the network, and allows the tracker to adapt to different night scenes. We also developed a transformer connection layer, which can help the network align cross-domain features. Secondly, we propose an efficient and lightweight attention block in the Transformer discriminator to increase frames per second while keeping the metrics as constant as possible, while the domain adaptive principle is used in the discriminator to form adversarial learning to reduce the domain differences in order to correctly distinguish the source domain (daytime) and the target domain (nighttime). Finally, we propose an adaptive hierarchical feature fusion module that efficiently fuses adjacent three-level feature information to facilitate the generation of unambiguous predictive tracking maps for subsequent input to the area regression network. We chose to test it on the NAT2021 public dataset and the UAVDark70 public dataset and tested three evaluation metrics that outperformed other state-of-the-art tracking methods. And our proposed tracker is seen to be very competitive in terms of correctness improvement of 4.93%, 8.26%, 9.42% and 8.49% in the NAT2021-test test set compared to SOTA tracking such as UDAT-CAR, UDAT-BAN, SparseTT and MixFormer.In addition, we introduced a new dataset called DarkBorder to validate the generalizability of our designed method when dealing with the task of UAV nighttime aerial video tracking. |
---|---|
AbstractList | In view of the complexity, uncertainty, and low visibility of the night-time environment. In this study, we designed an unsupervised domain adaptation framework, TransffCAR, using a Siamese network of the source (daytime) and target (nighttime) domains. Training patches are generated by finding targets from a large number of unlabelled frame images. This avoids heavy manual labelling, improves the performance of the network, and allows the tracker to adapt to different night scenes. We also developed a transformer connection layer, which can help the network align cross-domain features. Secondly, we propose an efficient and lightweight attention block in the Transformer discriminator to increase frames per second while keeping the metrics as constant as possible, while the domain adaptive principle is used in the discriminator to form adversarial learning to reduce the domain differences in order to correctly distinguish the source domain (daytime) and the target domain (nighttime). Finally, we propose an adaptive hierarchical feature fusion module that efficiently fuses adjacent three-level feature information to facilitate the generation of unambiguous predictive tracking maps for subsequent input to the area regression network. We chose to test it on the NAT2021 public dataset and the UAVDark70 public dataset and tested three evaluation metrics that outperformed other state-of-the-art tracking methods. And our proposed tracker is seen to be very competitive in terms of correctness improvement of 4.93%, 8.26%, 9.42% and 8.49% in the NAT2021-test test set compared to SOTA tracking such as UDAT-CAR, UDAT-BAN, SparseTT and MixFormer.In addition, we introduced a new dataset called DarkBorder to validate the generalizability of our designed method when dealing with the task of UAV nighttime aerial video tracking. |
Author | Zhao, Xueyi Guo, Rui Fang, Jian Wang, Deyong Wei, Haoran Fu, Yanyun |
Author_xml | – sequence: 1 givenname: Haoran orcidid: 0009-0000-2938-5299 surname: Wei fullname: Wei, Haoran organization: College of Computer Science and Technology, Xinjiang University, Ürümqi, China – sequence: 2 givenname: Yanyun surname: Fu fullname: Fu, Yanyun email: fyyun163@163.com organization: Beijing Academy of Science and Technology, Beijing, China – sequence: 3 givenname: Deyong surname: Wang fullname: Wang, Deyong organization: Key Laboratory of Big Data of Xinjiang Social Security Risk Prevention and Control, Xinjiang Lianhaichuangzhi Information Technology Company Ltd., Ürümqi, China – sequence: 4 givenname: Rui orcidid: 0000-0003-0659-4025 surname: Guo fullname: Guo, Rui organization: Key Laboratory of Big Data of Xinjiang Social Security Risk Prevention and Control, Xinjiang Lianhaichuangzhi Information Technology Company Ltd., Ürümqi, China – sequence: 5 givenname: Xueyi orcidid: 0009-0000-1827-0381 surname: Zhao fullname: Zhao, Xueyi organization: Key Laboratory of Big Data of Xinjiang Social Security Risk Prevention and Control, Xinjiang Lianhaichuangzhi Information Technology Company Ltd., Ürümqi, China – sequence: 6 givenname: Jian surname: Fang fullname: Fang, Jian organization: Key Laboratory of Big Data of Xinjiang Social Security Risk Prevention and Control, Xinjiang Lianhaichuangzhi Information Technology Company Ltd., Ürümqi, China |
BookMark | eNpNkMFOAyEQhompiVp9Aj3sC2yFhQX2WKtVE6MH9YwsDJXqLg2sGt9e1hpTLjP5h_-D-Y_QpA89IHRK8IwQ3JzPF4urx8dZhSs2o1RIQsQeOqwIb0paUz7Z6Q_QSUprnI_MUi0O0ctznz42ED99Alvc-9XrMPgOiod2DWYonqI2b75fFRd6nId-VPrkQuwgFrq3xWXotO-LudWbQQ8-31h-pLHcw_AV4tsx2nf6PcHJX52i5-XV0-KmvHu4vl3M70pTMS5K3kItjRGMSlrl_1mMG9zWOjcUc0ecbLQAy6CuqRGNbQgVzNomL80ZYY5O0e2Wa4Neq030nY7fKmivfoUQV0rHwZt3UJJYygUTmDnJJDNNW4OrRcutZi4_l1l0yzIxpBTB_fMIVmPmapu5GjNXf5ln19nW5QFgx8FERhL6AwrtfpM |
CODEN | IAECCG |
Cites_doi | 10.1609/aaai.v34i07.6944 10.1109/IROS51168.2021.9636309 10.1109/ICCV48922.2021.01028 10.1007/978-3-319-46448-0_27 10.1007/978-3-031-20047-2_1 10.1109/TITS.2020.3042973 10.1109/CVPR.2019.00441 10.1109/ICCV.2017.304 10.1007/978-3-030-58589-1_46 10.1109/ICCV48922.2021.01329 10.1007/s11263-020-01387-y 10.1109/CVPR42600.2020.00630 10.24963/ijcai.2022/127 10.1109/CVPR46437.2021.01350 10.1109/ICRA48506.2021.9561564 10.1109/CVPR52688.2022.00869 10.1109/ICCV.2019.00411 10.1109/LRA.2022.3146911 10.1109/CVPR.2018.00745 10.1109/CVPR.2009.5206848 10.1109/CVPR.2019.00087 10.1109/CVPR.2019.00472 10.48550/arXiv.2112.05561 10.1109/CVPR42600.2020.00716 10.1109/TPAMI.2019.2957464 10.1109/CVPR46437.2021.00810 10.1109/CVPR52688.2022.01324 10.1109/CVPR42600.2020.00670 |
ContentType | Journal Article |
DBID | 97E ESBDL RIA RIE AAYXX CITATION DOA |
DOI | 10.1109/ACCESS.2024.3378117 |
DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005-present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE Xplore CrossRef Open Access: DOAJ - Directory of Open Access Journals |
DatabaseTitle | CrossRef |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: DOA name: Open Access: DOAJ - Directory of Open Access Journals url: https://www.doaj.org/ sourceTypes: Open Website – sequence: 2 dbid: RIE name: IEEE Xplore url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering |
EISSN | 2169-3536 |
EndPage | 130913 |
ExternalDocumentID | oai_doaj_org_article_81d3674704f8484c9b5ef57b6da4f030 10_1109_ACCESS_2024_3378117 10473031 |
Genre | orig-research |
GrantInformation_xml | – fundername: Autonomous Region High-Level Talent Introduction Project: Research on UAV Information Reconnaissance and Big Data Analysis Technology for Public Safety grantid: U20B2060 – fundername: Key Laboratory of Big Data of Xinjiang Social Security Risk Prevention and Control – fundername: National Natural Science Foundation of China: Intelligent Perception and Real-Time Simulation and Deduction Technology for Urban Emergency Management Events grantid: U20B2060 |
GroupedDBID | 0R~ 4.4 5VS 6IK 97E AAJGR ACGFS ADBBV ALMA_UNASSIGNED_HOLDINGS BCNDV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD ESBDL GROUPED_DOAJ IFIPE IPLJI JAVBF KQ8 M43 M~E O9- OCL OK1 RIA RIE RIG RNS AAYXX CITATION |
ID | FETCH-LOGICAL-c2467-6be58cc743832000d0090b5a0d0306f1f89a7ed4e553c79d91374dd90246414f3 |
IEDL.DBID | RIE |
ISSN | 2169-3536 |
IngestDate | Mon Sep 30 19:22:44 EDT 2024 Wed Oct 02 14:38:24 EDT 2024 Wed Oct 02 05:56:46 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c2467-6be58cc743832000d0090b5a0d0306f1f89a7ed4e553c79d91374dd90246414f3 |
ORCID | 0009-0000-1827-0381 0009-0000-2938-5299 0000-0003-0659-4025 |
OpenAccessLink | https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/document/10473031 |
PageCount | 18 |
ParticipantIDs | crossref_primary_10_1109_ACCESS_2024_3378117 doaj_primary_oai_doaj_org_article_81d3674704f8484c9b5ef57b6da4f030 ieee_primary_10473031 |
PublicationCentury | 2000 |
PublicationDate | 20240000 2024-00-00 2024-01-01 |
PublicationDateYYYYMMDD | 2024-01-01 |
PublicationDate_xml | – year: 2024 text: 20240000 |
PublicationDecade | 2020 |
PublicationTitle | IEEE access |
PublicationTitleAbbrev | Access |
PublicationYear | 2024 |
Publisher | IEEE |
Publisher_xml | – name: IEEE |
References | ref13 ref12 ref15 ref14 ref53 Kingma (ref50) 2014 ref11 Liu (ref44) Kristan (ref52) Ye (ref6) 2022; 7 Chen (ref28) Müller (ref30) ref17 ref16 ref18 van der Maaten (ref46) 2008; 9 Wojke (ref22) ref45 ref48 ref47 ref42 ref41 ref43 Selvaraju (ref51) Bewley (ref21) ref49 ref8 ref4 ref3 ref5 Wu (ref9) Vaswani (ref10) 2017; 30 ref2 Zhu (ref24) Tzeng (ref35) ref1 Cao (ref25) ref38 Ye (ref7) Li (ref26) Li (ref39) 2022; 44 Wang (ref20) ref23 Ganin (ref33) 2016; 17 Cen (ref27) Huang (ref31) 2021; 43 Chen (ref37) Tzeng (ref36) Zhang (ref40) ref29 Fu (ref32) Ren (ref19); 28 Ganin (ref34); 2 |
References_xml | – ident: ref13 doi: 10.1609/aaai.v34i07.6944 – ident: ref14 doi: 10.1109/IROS51168.2021.9636309 – ident: ref11 doi: 10.1109/ICCV48922.2021.01028 – start-page: 3718 volume-title: Proc. 25th IEEE Int. Conf. Image Process. (ICIP) ident: ref27 article-title: Fully convolutional Siamese fusion networks for object tracking contributor: fullname: Cen – volume: 17 start-page: 2030 issue: 1 year: 2016 ident: ref33 article-title: Domain-adversarial training of neural networks publication-title: J. Mach. Learn. Res. contributor: fullname: Ganin – ident: ref53 doi: 10.1007/978-3-319-46448-0_27 – year: 2014 ident: ref50 article-title: Adam: A method for stochastic optimization publication-title: arXiv:1412.6980 contributor: fullname: Kingma – ident: ref23 doi: 10.1007/978-3-031-20047-2_1 – volume: 44 start-page: 4225 issue: 8 year: 2022 ident: ref39 article-title: Learning to enhance low-light image via zero-reference deep curve estimation publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TITS.2020.3042973 contributor: fullname: Li – start-page: 3339 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. ident: ref37 article-title: Domain adaptive faster R-CNN for object detection in the wild contributor: fullname: Chen – ident: ref4 doi: 10.1109/CVPR.2019.00441 – volume: 9 start-page: 2579 issue: 11 year: 2008 ident: ref46 article-title: Visualizing data using t-SNE publication-title: J. Mach. Learn. Res. contributor: fullname: van der Maaten – ident: ref48 doi: 10.1109/ICCV.2017.304 – start-page: 300 volume-title: Proc. Eur. Conf. Comput. Vis. (ECCV) ident: ref30 article-title: TrackingNet: A large-scale dataset and benchmark for object tracking in the wild contributor: fullname: Müller – start-page: 15437 volume-title: Proc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV) ident: ref25 article-title: HiFT: Hierarchical feature transformer for aerial tracking contributor: fullname: Cao – ident: ref16 doi: 10.1007/978-3-030-58589-1_46 – start-page: 2778 volume-title: Proc. IEEE/CVF Int. Conf. Comput. Vis. Workshops (ICCVW) ident: ref24 article-title: TPH-YOLOv5: Improved YOLOv5 based on transformer prediction head for object detection on drone-captured scenarios contributor: fullname: Zhu – ident: ref41 doi: 10.1109/ICCV48922.2021.01329 – start-page: 12146 volume-title: Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS) ident: ref32 article-title: HighlightNet: Highlighting low-light potential features for real-time UAV tracking contributor: fullname: Fu – ident: ref29 doi: 10.1007/s11263-020-01387-y – ident: ref3 doi: 10.1109/CVPR42600.2020.00630 – start-page: 1533 volume-title: Proc. IEEE/CVF Int. Conf. Comput. Vis. (ICCV) ident: ref40 article-title: Dynamic context-sensitive filtering network for video salient object detection contributor: fullname: Zhang – ident: ref2 doi: 10.24963/ijcai.2022/127 – start-page: 3464 volume-title: Proc. IEEE Int. Conf. Image Process. (ICIP) ident: ref21 article-title: Simple online and realtime tracking contributor: fullname: Bewley – start-page: 8122 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR) ident: ref28 article-title: Transformer tracking contributor: fullname: Chen – ident: ref45 doi: 10.1109/CVPR46437.2021.01350 – start-page: 3645 volume-title: Proc. IEEE Int. Conf. Image Process. (ICIP) ident: ref22 article-title: Simple online and realtime tracking with a deep association metric contributor: fullname: Wojke – ident: ref49 doi: 10.1109/ICRA48506.2021.9561564 – ident: ref5 doi: 10.1109/CVPR52688.2022.00869 – ident: ref17 doi: 10.1109/ICCV.2019.00411 – volume: 28 start-page: 1 volume-title: Proc. Adv. neural Inf. Process. Syst. ident: ref19 article-title: Faster R-CNN: Towards real-time object detection with region proposal networks contributor: fullname: Ren – volume: 7 start-page: 3866 issue: 2 year: 2022 ident: ref6 article-title: Tracker meets night: A transformer enhancer for UAV tracking publication-title: IEEE Robot. Autom. Lett. doi: 10.1109/LRA.2022.3146911 contributor: fullname: Ye – start-page: 2962 volume-title: Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR) ident: ref36 article-title: Adversarial discriminative domain adaptation contributor: fullname: Tzeng – ident: ref43 doi: 10.1109/CVPR.2018.00745 – start-page: 11999 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR) ident: ref44 article-title: Swin transformer v2: Scaling up capacity and resolution contributor: fullname: Liu – volume: 2 start-page: 1180 volume-title: Proc. 32nd Int. Conf. Mach. Learn. (ICML) ident: ref34 article-title: Unsupervised domain adaptation by backpropagation contributor: fullname: Ganin – ident: ref42 doi: 10.1109/CVPR.2009.5206848 – ident: ref38 doi: 10.1109/CVPR.2019.00087 – ident: ref18 doi: 10.1109/CVPR.2019.00472 – ident: ref47 doi: 10.48550/arXiv.2112.05561 – start-page: 15764 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR) ident: ref9 article-title: DANNet: A one-stage domain adaptation network for unsupervised nighttime semantic segmentation contributor: fullname: Wu – start-page: 618 volume-title: Proc. IEEE Int. Conf. Comput. Vis. (ICCV) ident: ref51 article-title: Grad-CAM: Visual explanations from deep networks via gradient-based localization contributor: fullname: Selvaraju – ident: ref15 doi: 10.1109/CVPR42600.2020.00716 – volume: 43 start-page: 1562 issue: 5 year: 2021 ident: ref31 article-title: GOT-10k: A large high-diversity benchmark for generic object tracking in the wild publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2019.2957464 contributor: fullname: Huang – ident: ref8 doi: 10.1109/CVPR46437.2021.00810 – start-page: 4068 volume-title: Proc. IEEE Int. Conf. Comput. Vis. (ICCV) ident: ref35 article-title: Simultaneous deep transfer across domains and tasks contributor: fullname: Tzeng – volume: 30 start-page: 1 year: 2017 ident: ref10 article-title: Attention is all you need publication-title: Adv. Neural Inf. Process. Syst. contributor: fullname: Vaswani – start-page: 1 volume-title: Proc. Eur. Conf. Comput. Vis. Workshops (ECCV) ident: ref52 article-title: The sixth visual object tracking VOT2018 challenge results contributor: fullname: Kristan – start-page: 7464 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR) ident: ref20 article-title: YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors contributor: fullname: Wang – start-page: 3079 volume-title: Proc. IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS) ident: ref7 article-title: DarkLighter: Light up the darkness for UAV tracking contributor: fullname: Ye – ident: ref1 doi: 10.1109/CVPR52688.2022.01324 – ident: ref12 doi: 10.1109/CVPR42600.2020.00670 – start-page: 8971 volume-title: Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. ident: ref26 article-title: High performance visual tracking with Siamese region proposal network contributor: fullname: Li |
SSID | ssj0000816957 |
Score | 2.3410816 |
Snippet | In view of the complexity, uncertainty, and low visibility of the night-time environment. In this study, we designed an unsupervised domain adaptation... |
SourceID | doaj crossref ieee |
SourceType | Open Website Aggregation Database Publisher |
StartPage | 130896 |
SubjectTerms | Adaptation models Feature extraction feature fusion Fuses Siamese network Single object tracking Target tracking Training transformer Transformers UAV Vectors |
SummonAdditionalLinks | – databaseName: Open Access: DOAJ - Directory of Open Access Journals dbid: DOA link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07T8MwELYQEwyIRxHlJQ-MlMa1HcdjW6gqhrK0Urfgp8RAWvXx_7lzUtROLGxREiXOnX333cX3HSFPvuBBxoJ1rPIQoARY6YYp1zGFDMLKzAmT2D4n-Xgm3udyvtfqC_eE1fTAteC6gKd4Dpg3E7EQhXDayhClsrk3IsIMTdaXyb1gKtngguVaqoZmiGW62x8O4YsgIOyJF85TfeWBK0qM_QctVpKHGZ2TswYa0n49pAtyFKpLcrpHGHhFPmfVervE5b0Onk4wsMbm8PTDYjqFguNxmPqmA4PXFxWd7nBpWFFTefq6-DZfFe17s6z_wdPRFvNldFJvB2-R2ehtOhx3mh4JHddDG5fbIAvnFDKOYtWNB8yUWWngAIKByGKhjQpeBCm5U9prxpXwXoMYcsFE5NfkuFpU4YZQC2jI6l7kHgvkkdkuxAjeu2ej8o6bNnneiatc1lQYZQohMl3W0i1RumUj3TYZoEh_b0Ue63QCtFs22i3_0m6btFAhe-8TYJI4u_2Ph9-RExxwnVC5J8eb1TY8AMTY2Mc0m34AGcHKag priority: 102 providerName: Directory of Open Access Journals |
Title | Unsupervised Nighttime Object Tracking Based on Transformer and Domain Adaptation Fusion Network |
URI | https://ieeexplore.ieee.org/document/10473031 https://doaj.org/article/81d3674704f8484c9b5ef57b6da4f030 |
Volume | 12 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV07T8MwELaACQbeiPKSB0ZaktiO47EUKsRQFiqxBT8lhEgraBZ-PXd2igAJic1yItm-8-O-s-87Qs5dxbwIVd430gFA8bDSdS5tX1fCcyMyy3Vk-5yUt1N-9ygeu2D1GAvjvY-Pz_wAi_Eu381si66yS6QVgC0XwM5qlRUpWOvLoYIZJJSQHbNQnqnL4WgEgwAMWPABYzGk8sfpE0n6f2RViYfKeItMlt1Jb0leBu3CDOzHL6bGf_d3m2x25iUdpvmwQ1Z8s0s2vpEO7pGnafPeznGLePeOThCcY4J5em_QJUPh8LLoPqdXGr_PGvqwtG39G9WNo9ezV_3c0KHT83SPT8ct-tzoJD0p3yfT8c3D6Lbf5Vno2wL3ydJ4UVkrkbUUI3cc2F2ZERoKAChCHiqlpXfcC8GsVE7lTHLnFMi15DkP7ICsNbPGHxJqwKIyqgjMYZA9suP5EMACKEyQzjLdIxdL-dfzRKdRRxiSqTqpq0Z11Z26euQKdfT1K3JhxwoQc90trRosblYCKsp4qHjFrTLQqJCmdJoHGEGP7KNqvrWXtHL0R_0xWcc-JD_LCVlbvLX-FCyPhTmLiP0szrtPxFfV0A |
link.rule.ids | 315,786,790,802,870,2115,4043,27956,27957,27958,55109 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1LTxsxELYqeigc-uChBij1gWMTdmN7vT6GQJQW2F4Sidvip4SqbiLIXvj1zNgbBEhI3CzvSn6M7flm7PmGkGNXMi9CmfeNdGCgeNjpOpe2r0vhuRGZ5TqyfVbFdM7_XIvrLlg9xsJ47-PjMz_AYrzLdwvboqvsBGkF4MgFY-cjKPpMpXCtJ5cK5pBQQnbcQvD9ZDQewzDAChzyAWMxqPKF_ok0_S_yqkS1MvlCqnWH0muSf4N2ZQb24RVX47t7_JV87gAmHaUV8Y188M022XpGO7hDbubNfbvEQ-LeO1qheY4p5ulfg04ZCurLogOdnmr8vmjobI1u_R3VjaNni__6tqEjp5fpJp9OWvS60So9Kt8l88n5bDztd5kW-naIJ2VhvCitlchbirE7DpBXZoSGApgUIQ-l0tI77oVgViqncia5cwrmteA5D2yPbDSLxn8n1ACmMmoYmMMwe-TH8yEABhiaIJ1lukd-ree_XiZCjToaIpmqk7hqFFfdiatHTlFGT78iG3asgGmuu81VA-ZmBdhFGQ8lL7lVBhoV0hRO8wAj6JFdFM2z9pJU9t-o_0k-TWdXl_Xl7-rigGxif5LX5ZBsrO5a_wNwyMocxdX3CCrq2DE |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Unsupervised+Nighttime+Object+Tracking+Based+on+Transformer+and+Domain+Adaptation+Fusion+Network&rft.jtitle=IEEE+access&rft.au=Wei%2C+Haoran&rft.au=Fu%2C+Yanyun&rft.au=Wang%2C+Deyong&rft.au=Guo%2C+Rui&rft.date=2024&rft.pub=IEEE&rft.eissn=2169-3536&rft.volume=12&rft.spage=130896&rft.epage=130913&rft_id=info:doi/10.1109%2FACCESS.2024.3378117&rft.externalDocID=10473031 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2169-3536&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2169-3536&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2169-3536&client=summon |