Facing the Void: Overcoming Missing Data in Multi-View Imagery
In some scenarios, a single input image may not be enough to allow the object classification. In those cases, it is crucial to explore the complementary information extracted from images presenting the same object from multiple perspectives (or views) in order to enhance the general scene understand...
Saved in:
Published in | IEEE access Vol. 11; pp. 12546 - 12553 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
IEEE
2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | In some scenarios, a single input image may not be enough to allow the object classification. In those cases, it is crucial to explore the complementary information extracted from images presenting the same object from multiple perspectives (or views) in order to enhance the general scene understanding and, consequently, increase the performance. However, this task, commonly called multi-view image classification, has a major challenge: missing data. In this paper, we propose a novel technique for multi-view image classification robust to this problem. The proposed method, based on state-of-the-art deep learning-based approaches and metric learning, can be easily adapted and exploited in other applications and domains. A systematic evaluation of the proposed algorithm was conducted using two multi-view aerial-ground datasets with very distinct properties. Results show that the proposed algorithm provides improvements in multi-view image classification accuracy when compared to state-of-the-art methods. The code of the proposed approach is available at https://github.com/Gabriellm2003/remote_sensing_missing_data . |
---|---|
AbstractList | In some scenarios, a single input image may not be enough to allow the object classification. In those cases, it is crucial to explore the complementary information extracted from images presenting the same object from multiple perspectives (or views) in order to enhance the general scene understanding and, consequently, increase the performance. However, this task, commonly called multi-view image classification, has a major challenge: missing data. In this paper, we propose a novel technique for multi-view image classification robust to this problem. The proposed method, based on state-of-the-art deep learning-based approaches and metric learning, can be easily adapted and exploited in other applications and domains. A systematic evaluation of the proposed algorithm was conducted using two multi-view aerial-ground datasets with very distinct properties. Results show that the proposed algorithm provides improvements in multi-view image classification accuracy when compared to state-of-the-art methods. The code of the proposed approach is available at https://github.com/Gabriellm2003/remote_sensing_missing_data . |
Author | Nogueira, Keiller Santos, Jefersson A. Dos Machado, Gabriel Pereira, Matheus B. |
Author_xml | – sequence: 1 givenname: Gabriel orcidid: 0000-0002-7133-6324 surname: Machado fullname: Machado, Gabriel email: gabriel.lucas@dcc.ufmg.br organization: Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil – sequence: 2 givenname: Matheus B. orcidid: 0000-0002-2471-2358 surname: Pereira fullname: Pereira, Matheus B. organization: Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil – sequence: 3 givenname: Keiller orcidid: 0000-0003-3308-6384 surname: Nogueira fullname: Nogueira, Keiller organization: Department of Computing Science and Mathematics, University of Stirling, Stirling, Scotland, U.K – sequence: 4 givenname: Jefersson A. Dos orcidid: 0000-0002-8889-1586 surname: Santos fullname: Santos, Jefersson A. Dos organization: Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil |
BookMark | eNpNkNFOwjAUhhuDiYg8ATd7gWFPu3WrFyZkipJAuEC5bbr2DEtgM93U8PZujhDPzX_yJ99_8d2SQVmVSMgE6BSAyvtZlj1vNlNGGZtyxkFAckWGDIQMeczF4N9_Q8Z1vaftpW0VJ0PyONfGlbug-cBgWzn7EKy_0Zvq2JUrV9ddPulGB64MVl-HxoVbhz_B4qh36E935LrQhxrH5xyR9_nzW_YaLtcvi2y2DA1PaRMCcgYaWGJFHmECuQARG7BAIy5AFkALnlspTYpUC9QMAEUec7QRTxhqPiKLftdWeq8-vTtqf1KVduqvqPxOad84c0AlTW4jSDSXEiPZwmjTmInc5qI1Y027xfst46u69lhc9oCqzqjqjarOqDobbalJTzlEvBBSyiRuzf4CEsNyTA |
CODEN | IAECCG |
CitedBy_id | crossref_primary_10_3390_jimaging10040094 crossref_primary_10_1016_j_rse_2024_114112 |
Cites_doi | 10.1007/978-3-030-01231-1_5 10.1016/j.rse.2019.04.014 10.3390/rs11111259 10.1109/TPAMI.2018.2798607 10.1109/CVPR.2018.00758 10.1109/CVPR.2016.90 10.1109/TKDE.2018.2872063 10.1145/3209978.3210036 10.1109/TKDE.2018.2791607 10.1109/JSTARS.2020.3033424 10.1109/CVPR.2019.00060 10.1145/3219819.3219963 10.1162/0899766042321814 10.1016/j.inffus.2019.02.010 10.1109/SSD52085.2021.9429478 10.1109/CVPR.2009.5206848 10.1371/journal.pone.0245230 10.1016/j.inffus.2021.06.008 10.1109/CVPR42600.2020.00412 10.1007/s10791-020-09377-x 10.1109/CVPR.2017.243 10.1145/3422622 10.1007/978-3-319-46448-0_30 10.1007/978-3-319-24574-4_78 10.1109/CVPR.2019.00259 |
ContentType | Journal Article |
DBID | 97E ESBDL RIA RIE AAYXX CITATION DOA |
DOI | 10.1109/ACCESS.2022.3231617 |
DatabaseName | IEEE All-Society Periodicals Package (ASPP) 2005–Present IEEE Xplore Open Access Journals IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Xplore Digital Library CrossRef DOAJ Directory of Open Access Journals |
DatabaseTitle | CrossRef |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: DOA name: 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 | 12553 |
ExternalDocumentID | oai_doaj_org_article_9cbd417a399e492eaed8526bdb6617dc 10_1109_ACCESS_2022_3231617 9997535 |
Genre | orig-research |
GrantInformation_xml | – fundername: Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) grantid: APQ-00449-17 funderid: 10.13039/501100004901 – fundername: Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) grantid: 306955/2021-0 funderid: 10.13039/501100003593 – fundername: Serrapilheira Institute grantid: Serra—R-2011-37776 funderid: 10.13039/501100013275 – fundername: Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)—Finance Code 001 funderid: 10.13039/501100002322 |
GroupedDBID | 0R~ 4.4 5VS 6IK 97E AAJGR ABVLG 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-c380t-1e321a127d6b4e71b6165c1d1043619f10f3bd99c8e0a6ea211e6b53ed4372ea3 |
IEDL.DBID | RIE |
ISSN | 2169-3536 |
IngestDate | Thu Sep 05 15:41:46 EDT 2024 Fri Aug 23 03:12:41 EDT 2024 Wed Jun 26 19:28:10 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c380t-1e321a127d6b4e71b6165c1d1043619f10f3bd99c8e0a6ea211e6b53ed4372ea3 |
ORCID | 0000-0002-2471-2358 0000-0003-3308-6384 0000-0002-7133-6324 0000-0002-8889-1586 |
OpenAccessLink | https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/document/9997535 |
PageCount | 8 |
ParticipantIDs | ieee_primary_9997535 crossref_primary_10_1109_ACCESS_2022_3231617 doaj_primary_oai_doaj_org_article_9cbd417a399e492eaed8526bdb6617dc |
PublicationCentury | 2000 |
PublicationDate | 20230000 2023-00-00 2023-01-01 |
PublicationDateYYYYMMDD | 2023-01-01 |
PublicationDate_xml | – year: 2023 text: 20230000 |
PublicationDecade | 2020 |
PublicationTitle | IEEE access |
PublicationTitleAbbrev | Access |
PublicationYear | 2023 |
Publisher | IEEE |
Publisher_xml | – name: IEEE |
References | ref13 ref12 ref15 ref14 Kingma (ref27) 2014 ref11 ref10 ref2 ref1 ref17 ref19 ref18 Simonyan (ref24) 2014 Shi (ref16); 32 Caron (ref22); 33 ref23 ref26 ref25 ref20 ref21 ref28 ref29 ref8 ref7 ref9 ref4 ref3 ref6 ref5 |
References_xml | – ident: ref14 doi: 10.1007/978-3-030-01231-1_5 – ident: ref5 doi: 10.1016/j.rse.2019.04.014 – ident: ref3 doi: 10.3390/rs11111259 – ident: ref21 doi: 10.1109/TPAMI.2018.2798607 – year: 2014 ident: ref27 article-title: Adam: A method for stochastic optimization publication-title: arXiv:1412.6980 contributor: fullname: Kingma – ident: ref15 doi: 10.1109/CVPR.2018.00758 – ident: ref23 doi: 10.1109/CVPR.2016.90 – ident: ref2 doi: 10.1109/TKDE.2018.2872063 – volume: 32 start-page: 1 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref16 article-title: Spatial-aware feature aggregation for image based cross-view geo-localization contributor: fullname: Shi – ident: ref19 doi: 10.1145/3209978.3210036 – ident: ref7 doi: 10.1109/TKDE.2018.2791607 – ident: ref20 doi: 10.1109/JSTARS.2020.3033424 – volume: 33 start-page: 9912 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref22 article-title: Unsupervised learning of visual features by contrasting cluster assignments contributor: fullname: Caron – ident: ref26 doi: 10.1109/CVPR.2019.00060 – ident: ref8 doi: 10.1145/3219819.3219963 – ident: ref12 doi: 10.1162/0899766042321814 – ident: ref9 doi: 10.1016/j.inffus.2019.02.010 – ident: ref11 doi: 10.1109/SSD52085.2021.9429478 – ident: ref29 doi: 10.1109/CVPR.2009.5206848 – ident: ref1 doi: 10.1371/journal.pone.0245230 – ident: ref6 doi: 10.1016/j.inffus.2021.06.008 – ident: ref17 doi: 10.1109/CVPR42600.2020.00412 – ident: ref28 doi: 10.1007/s10791-020-09377-x – ident: ref25 doi: 10.1109/CVPR.2017.243 – ident: ref13 doi: 10.1145/3422622 – ident: ref18 doi: 10.1007/978-3-319-46448-0_30 – ident: ref4 doi: 10.1007/978-3-319-24574-4_78 – year: 2014 ident: ref24 article-title: Very deep convolutional networks for large-scale image recognition publication-title: arXiv:1409.1556 contributor: fullname: Simonyan – ident: ref10 doi: 10.1109/CVPR.2019.00259 |
SSID | ssj0000816957 |
Score | 2.3383863 |
Snippet | In some scenarios, a single input image may not be enough to allow the object classification. In those cases, it is crucial to explore the complementary... |
SourceID | doaj crossref ieee |
SourceType | Open Website Aggregation Database Publisher |
StartPage | 12546 |
SubjectTerms | cross-view matching Data mining Data models Feature extraction Image classification metric learning multi-modal machine learning multi-view missing data completion Remote sensing Task analysis |
SummonAdditionalLinks | – databaseName: DOAJ Directory of Open Access Journals dbid: DOA link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV1LS8NAEF6kJz2IWsX6IgePxmaf6XoQarVUQb3Y0lvYxwR6MBWJiP_e2U3UevLiKcuyDMk3w8432eQbQk5Z7jVHGp7SPLep8FCmxnOeavBIZo1Sxka1zwc1mYq7uZyvtPoK34Q18sANcH3trBc0N5hIQWgGBo1Ipqy3mFly7-LuS-VKMRX34AFVWuatzBDNdH84GuETYUHI2DlHUqNii7KfVBQV-3-1WIkZZrxFNltqmAybW9oma1DtkI0VwcAuuRwbh4MEWVsyWy78RfKIoYhBEybvEcNwvTa1SRZVEv-tTWcLeE9un4NUxccumY5vnkaTtO2AkDo-yOqUAmfUUARUWQE5tYoq6ainQTie6pJmJbdeazeAzCgwiAQoKzn4cBwHhu-RTrWsYJ8kEJTbwHGlSi986SxzzkmWeS_KTHjaI2dfYBQvjdBFEQuETBcNdkXArmix65GrANj30qBSHSfQd0Xru-Iv3_VIN8D9bQTJKlZP8uA_bB-S9dAZvnlbckQ69esbHCN_qO1JDJVP7njANg priority: 102 providerName: Directory of Open Access Journals |
Title | Facing the Void: Overcoming Missing Data in Multi-View Imagery |
URI | https://ieeexplore.ieee.org/document/9997535 https://doaj.org/article/9cbd417a399e492eaed8526bdb6617dc |
Volume | 11 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1RT9swED5Bn-CBMQqiG1R52CNp4zhx6j1Mgm4VmwS80Kpvke27SBVaO02ppu3X7-yEDBAPe4plWdHl7iJ_d_Z9B_AhLVBLhuGxKAobZ0hVbFDKWBMymDVKGRvYPm_V9Tz7tsyXO3DR1cIQUbh8RiM_DGf5uHFbnyobM5hhdJ3vwu4kSZtarS6f4htI6LxoiYVEoseX0yl_A4eAaTqSDGNUaEr2b_MJHP3PmqqEPWX2Bm4epWmukjyMtrUduT8viBr_V9xDOGjBZXTZeMNb2KH1Eew_oRzsw6eZcTyIGPdFi80KP0Z37Mzsdn7yhq3gn59NbaLVOgrVufFiRb-ir9892cXvY5jPvtxPr-O2h0Ls5CSpY0EyFUawSZTNqBBWCZU7gcJTzwtdiaSSFrV2E0qMIsPxICmbS0J_oEdGnkBvvVnTKUTkud_ISaUqzLByNnXO5WmCmFVJhmIAF4_KLX80VBllCDESXTa2KL0tytYWA7jyBuiWep7rMME6LNvfptTOYiYKwzCKMs3ysAvlqbJoGVcU6AbQ93rvXtKq_N3r0-9hz3eLbzIoZ9Crf27pnDFFbYchFh8Gl_oLZIbJ2w |
link.rule.ids | 315,786,790,802,870,2115,4043,27954,27955,27956,55107 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1LT9wwEB4BPUAPbXmJLS3kwJEscZw46x6Q6JbV8li4AOIW2Z6JtELsVlVWFfz6jp2wQMWBUyzLiibzTeRv_PgGYC8tUEum4bEoChtnSFVsUMpYEzKZNUoZG9Q-L9TwOju9zW8XYH9-F4aIwuEz6vpm2MvHqZv5pbIDJjPMrvNF-MDzfFI0t7XmKyq-hITOi1ZaSCT64Kjf56_gJDBNu5KJjAplyZ6nn6DS_6qsSphVBp9h9GRPc5jkrjurbdc9_ifV-F6Dv8Cnll5GR008rMICTdbg4wvRwXU4HBjHjYiZX3QzHeOP6JLDmQPPd44YB__8ZWoTjSdRuJ8b34zpb3Ry7-UuHjbgenB81R_GbRWF2MleUseCZCqMYFCUzagQVgmVO4HCi88LXYmkkha1dj1KjCLDGSEpm0tCv6VHRm7C0mQ6oS2IyKu_kZNKVZhh5WzqnMvTBDGrkgxFB_afnFv-bsQyypBkJLpssCg9FmWLRQd-egDmQ73SdehgH5btj1NqZzEThWEiRZlmeziI8lRZtMwsCnQdWPd-n7-kdfnXt7t3YXl4NTovz08uzrZhxdeOb9ZTvsFS_WdG35lh1HYnBNY_bpjMOg |
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=Facing+the+Void%3A+Overcoming+Missing+Data+in+Multi-View+Imagery&rft.jtitle=IEEE+access&rft.au=Machado%2C+Gabriel&rft.au=Pereira%2C+Matheus+B.&rft.au=Nogueira%2C+Keiller&rft.au=Santos%2C+Jefersson+A.+Dos&rft.date=2023&rft.issn=2169-3536&rft.eissn=2169-3536&rft.volume=11&rft.spage=12547&rft.epage=12554&rft_id=info:doi/10.1109%2FACCESS.2022.3231617&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_ACCESS_2022_3231617 |
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 |