Encoder-Decoder With Cascaded CRFs for Semantic Segmentation

When dealing with semantic segmentation, how to locate the object boundary information more accurately is a key problem to distinguish different objects better. The existing methods lose some image information more or less in the process of feature extraction, which also includes the boundary and co...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transactions on circuits and systems for video technology Ročník 31; číslo 5; s. 1926 - 1938
Hlavní autoři:	Ji, Jian, Shi, Rui, Li, Sitong, Chen, Peng, Miao, Qiguang
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York IEEE 01.05.2021 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	boundary location Coders Computer vision Conditional random fields Convolution Decoding encoder-decoder Encoders-Decoders Feature extraction fully convolution network Image segmentation Multilayers Semantic segmentation Semantics Task analysis
ISSN:	1051-8215, 1558-2205
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	When dealing with semantic segmentation, how to locate the object boundary information more accurately is a key problem to distinguish different objects better. The existing methods lose some image information more or less in the process of feature extraction, which also includes the boundary and context information. At present, some semantic segmentation methods use CRFs (conditional random fields) to obtain boundary information, but they usually only deal with the final output of the model. In this article, inspired by the skip connection of FCN (Fully convolution network) and the good boundary refinement ability of CRFs, a cascaded CRFs is designed and introduced into the decoder of semantic segmentation model to learn boundary information from multi-layers and enhance the ability of the model in object boundary location. Furthermore, in order to supplement the semantic information of images, the output of the cascaded CRFs is fused with the output of the last decoder, so that the model can enhance the ability of locating the object boundary and get more accurate semantic segmentation results. Finally, a number of experiments on different datasets illustrate the feasibility and efficiency of our method, showing that our method enhances the model's ability to locate target boundary information.
AbstractList	When dealing with semantic segmentation, how to locate the object boundary information more accurately is a key problem to distinguish different objects better. The existing methods lose some image information more or less in the process of feature extraction, which also includes the boundary and context information. At present, some semantic segmentation methods use CRFs (conditional random fields) to obtain boundary information, but they usually only deal with the final output of the model. In this article, inspired by the skip connection of FCN (Fully convolution network) and the good boundary refinement ability of CRFs, a cascaded CRFs is designed and introduced into the decoder of semantic segmentation model to learn boundary information from multi-layers and enhance the ability of the model in object boundary location. Furthermore, in order to supplement the semantic information of images, the output of the cascaded CRFs is fused with the output of the last decoder, so that the model can enhance the ability of locating the object boundary and get more accurate semantic segmentation results. Finally, a number of experiments on different datasets illustrate the feasibility and efficiency of our method, showing that our method enhances the model’s ability to locate target boundary information.
Author	Li, Sitong Miao, Qiguang Shi, Rui Chen, Peng Ji, Jian
Author_xml	– sequence: 1 givenname: Jian orcidid: 0000-0001-6427-6725 surname: Ji fullname: Ji, Jian email: jji@xidian.edu.cn organization: School of Computer Science and Technology, Xidian University, Xi'an, China – sequence: 2 givenname: Rui surname: Shi fullname: Shi, Rui organization: School of Computer Science and Technology, Xidian University, Xi'an, China – sequence: 3 givenname: Sitong orcidid: 0000-0002-2950-0047 surname: Li fullname: Li, Sitong organization: School of Computer Science and Technology, Xidian University, Xi'an, China – sequence: 4 givenname: Peng orcidid: 0000-0002-1669-0013 surname: Chen fullname: Chen, Peng organization: School of Computer Science and Technology, Xidian University, Xi'an, China – sequence: 5 givenname: Qiguang orcidid: 0000-0001-6766-8310 surname: Miao fullname: Miao, Qiguang email: qgmiao@xidian.edu.cn organization: School of Computer Science and Technology, Xidian University, Xi'an, China
BookMark	eNp9kDtPwzAUhS1UJNrCH4AlEnOKH7FjSywotIBUCYkWGK3rRyBVmxQ7Hfj3pE3FwMB0znC-e3TPCA3qpvYIXRI8IQSrm2WxeFtOKKZ4wjDhUogTNCScy5RSzAedx5ykkhJ-hkYxrjAmmczyIbqd1rZxPqT3_qDJe9V-JgVEC867pHiZxaRsQrLwG6jbynbmY-PrFtqqqc_RaQnr6C-OOkavs-myeEznzw9Pxd08tVTxNnVMABfSKYVdbjOwwsjcWKMUyxiR3uQWjALBSwouM4Jy4CWAcd5goESxMbru725D87XzsdWrZhfqrlJTTiklWY5Zl6J9yoYmxuBLvQ3VBsK3JljvV9KHlfR-JX1cqYPkH8hW_XNtgGr9P3rVo5X3_rdLEcEpE-wHSDZ21Q
CODEN	ITCTEM
CitedBy_id	crossref_primary_10_1109_TAI_2024_3363685 crossref_primary_10_1109_TCSVT_2022_3179021 crossref_primary_10_1109_TCSVT_2024_3367370 crossref_primary_10_1109_TCSVT_2024_3427720 crossref_primary_10_1109_TCSVT_2024_3454227 crossref_primary_10_1109_TCSVT_2023_3285091 crossref_primary_10_1016_j_engappai_2022_105431 crossref_primary_10_1109_TCDS_2022_3225200 crossref_primary_10_4218_etrij_2023_0017 crossref_primary_10_1109_TCSVT_2023_3298644 crossref_primary_10_1109_TCSVT_2024_3509504 crossref_primary_10_1109_TCSVT_2022_3218076 crossref_primary_10_1007_s00371_025_03798_9 crossref_primary_10_1007_s00530_025_01888_1 crossref_primary_10_1109_TCSVT_2024_3441036 crossref_primary_10_1038_s41598_024_74690_0 crossref_primary_10_1109_TCSVT_2023_3317937 crossref_primary_10_1007_s00371_022_02477_3 crossref_primary_10_1109_TCSVT_2022_3193612 crossref_primary_10_1016_j_patrec_2023_12_023 crossref_primary_10_1016_j_compbiomed_2025_110851 crossref_primary_10_1109_TIM_2022_3224526 crossref_primary_10_1016_j_eswa_2024_124019 crossref_primary_10_1109_TCSVT_2024_3412996 crossref_primary_10_1016_j_dsp_2025_105264 crossref_primary_10_1109_TCSVT_2023_3292156 crossref_primary_10_1109_TCSVT_2024_3464631 crossref_primary_10_1109_TCSVT_2022_3146305 crossref_primary_10_1109_TCSVT_2022_3212088 crossref_primary_10_1109_TCSVT_2024_3375789 crossref_primary_10_1109_TCSVT_2023_3293166 crossref_primary_10_1007_s00530_024_01429_2 crossref_primary_10_1109_TCSVT_2021_3121680 crossref_primary_10_1016_j_heliyon_2024_e31626 crossref_primary_10_1109_TCSVT_2022_3208833 crossref_primary_10_1016_j_neucom_2024_128632 crossref_primary_10_1109_TCSVT_2023_3325629 crossref_primary_10_1109_TCSVT_2023_3343632 crossref_primary_10_1016_j_neucom_2024_128991 crossref_primary_10_1109_TCSVT_2022_3229359 crossref_primary_10_1109_TIV_2023_3314527 crossref_primary_10_1109_TCSVT_2023_3252807 crossref_primary_10_1109_TCSVT_2022_3206496 crossref_primary_10_1109_TCSVT_2023_3265075 crossref_primary_10_1109_TCSVT_2021_3138795 crossref_primary_10_1109_TCSVT_2022_3187664 crossref_primary_10_1109_TCSVT_2024_3462946 crossref_primary_10_4018_IJVAR_307063 crossref_primary_10_1109_TCSVT_2025_3536086 crossref_primary_10_1016_j_displa_2025_103168 crossref_primary_10_1109_TCSVT_2025_3559084 crossref_primary_10_1109_TCSVT_2024_3413778 crossref_primary_10_1109_TCSVT_2024_3391304 crossref_primary_10_1109_TCSVT_2023_3296162 crossref_primary_10_1109_TCSVT_2023_3275314 crossref_primary_10_1109_TCSVT_2024_3511926 crossref_primary_10_1016_j_cviu_2025_104411 crossref_primary_10_1109_TCSVT_2020_3037234 crossref_primary_10_1007_s11760_023_02777_8 crossref_primary_10_1109_TCSVT_2022_3171968 crossref_primary_10_1109_TCSVT_2024_3435045 crossref_primary_10_1109_TCSVT_2021_3109084 crossref_primary_10_1109_TCSVT_2023_3239541 crossref_primary_10_1109_TCSVT_2021_3056726 crossref_primary_10_1109_TCSVT_2023_3236432 crossref_primary_10_1109_TCSVT_2023_3300846 crossref_primary_10_1109_TCSVT_2023_3325360 crossref_primary_10_1016_j_engappai_2023_107642 crossref_primary_10_1109_TCSVT_2024_3502837 crossref_primary_10_1007_s40747_023_01063_x crossref_primary_10_1109_TCSVT_2023_3298796 crossref_primary_10_1109_TCSVT_2022_3206476 crossref_primary_10_1142_S0129065725500388 crossref_primary_10_1109_TCSVT_2022_3231964 crossref_primary_10_1007_s40747_023_01103_6 crossref_primary_10_1109_TCSVT_2021_3126591 crossref_primary_10_1109_TCSVT_2023_3339683 crossref_primary_10_1109_TCSVT_2022_3205182 crossref_primary_10_1109_TCSVT_2022_3170048 crossref_primary_10_1016_j_engappai_2024_109294 crossref_primary_10_1109_TCSVT_2023_3284076 crossref_primary_10_1109_TCSVT_2023_3260240 crossref_primary_10_1016_j_neucom_2025_130504
Cites_doi	10.1109/TPAMI.2019.2938758 10.1109/CVPR.2018.00759 10.1109/CVPR.2018.00255 10.1109/ICCV.2015.203 10.1109/CVPR.2017.690 10.1007/978-3-030-58604-1_20 10.1109/TPAMI.2016.2572683 10.1109/CVPR.2016.90 10.1109/TPAMI.2015.2389824 10.1109/TPAMI.2011.131 10.1016/j.patrec.2008.04.005 10.1109/ICCV.2017.533 10.1109/CVPR.2017.25 10.1109/TPAMI.2016.2644615 10.1109/ICCV.2015.178 10.1109/TPAMI.2017.2699184 10.1103/RevModPhys.54.235 10.1109/CVPR.2017.660 10.1109/CVPR.2017.243 10.1109/CVPR.2018.00256 10.1007/978-3-030-01249-6_34 10.1109/CVPR.2019.00017 10.1109/TPAMI.2018.2852750 10.1109/TGRS.2018.2815613 10.1007/978-3-642-33715-4_54 10.1109/TIM.2020.3004902 10.1109/TPAMI.2017.2708714 10.1007/s11263-007-0090-8 10.1109/CVPR.2017.549 10.1109/ICCV.2019.00216 10.1109/ICCV.2015.179 10.1109/CVPR.2013.79 10.1109/TMM.2015.2477035
ContentType	Journal Article
Copyright	Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021
Copyright_xml	– notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021
DBID	97E RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D
DOI	10.1109/TCSVT.2020.3015866
DatabaseName	IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE/IET Electronic Library CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional
DatabaseTitle	CrossRef Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional
DatabaseTitleList	Technology Research Database
Database_xml	– sequence: 1 dbid: RIE name: IEEE/IET Electronic Library url: https://ieeexplore.ieee.org/ sourceTypes: Publisher
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering
EISSN	1558-2205
EndPage	1938
ExternalDocumentID	10_1109_TCSVT_2020_3015866 9165236
Genre	orig-research
GroupedDBID	-~X 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AARMG AASAJ AAWTH ABAZT ABQJQ ABVLG ACGFO ACGFS ACIWK AENEX AETIX AGQYO AGSQL AHBIQ AI. AIBXA AKJIK AKQYR ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 EBS EJD HZ~ H~9 ICLAB IFIPE IFJZH IPLJI JAVBF LAI M43 O9- OCL P2P RIA RIE RNS RXW TAE TN5 VH1 AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D
ID	FETCH-LOGICAL-c295t-d36a568d990d7c4ac6b87bcb9934318eb7cab9a65f2ad4b625a5faabdeb0a2193
IEDL.DBID	RIE
ISICitedReferencesCount	83
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000647394100020&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1051-8215
IngestDate	Sun Nov 09 06:31:36 EST 2025 Sat Nov 29 01:44:14 EST 2025 Tue Nov 18 22:27:54 EST 2025 Wed Aug 27 02:30:02 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	5
Language	English
License	https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c295t-d36a568d990d7c4ac6b87bcb9934318eb7cab9a65f2ad4b625a5faabdeb0a2193
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ORCID	0000-0002-2950-0047 0000-0002-1669-0013 0000-0001-6427-6725 0000-0001-6766-8310
PQID	2522214703
PQPubID	85433
PageCount	13
ParticipantIDs	crossref_citationtrail_10_1109_TCSVT_2020_3015866 crossref_primary_10_1109_TCSVT_2020_3015866 ieee_primary_9165236 proquest_journals_2522214703
PublicationCentury	2000
PublicationDate	2021-05-01
PublicationDateYYYYMMDD	2021-05-01
PublicationDate_xml	– month: 05 year: 2021 text: 2021-05-01 day: 01
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	IEEE transactions on circuits and systems for video technology
PublicationTitleAbbrev	TCSVT
PublicationYear	2021
Publisher	IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Publisher_xml	– name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
References	ref35 ref12 hazirbas (ref22) 2016 ref37 ref36 ref14 ref31 ref30 ref33 ref11 chen (ref13) 2014 ref32 ref10 ref2 ref1 ref39 ref17 ref38 ref16 ref19 ref18 ref24 ref23 liu (ref9) 2020 ref26 ref25 ref20 ref21 chen (ref7) 2017 ref28 ref27 ref29 ref8 hou (ref34) 2018 yuan (ref15) 2019 ref4 ref3 ref6 ref5
References_xml	– ident: ref29 doi: 10.1109/TPAMI.2019.2938758 – ident: ref35 doi: 10.1109/CVPR.2018.00759 – year: 2019 ident: ref15 article-title: Object-contextual representations for semantic segmentation publication-title: arXiv 1909 11065 – ident: ref28 doi: 10.1109/CVPR.2018.00255 – start-page: 213 year: 2016 ident: ref22 article-title: FuseNet: Incorporating depth into semantic segmentation via fusion-based cnn architecture publication-title: Proc Asian Conf Comput Vis – ident: ref31 doi: 10.1109/ICCV.2015.203 – ident: ref3 doi: 10.1109/CVPR.2017.690 – ident: ref16 doi: 10.1007/978-3-030-58604-1_20 – ident: ref4 doi: 10.1109/TPAMI.2016.2572683 – year: 2020 ident: ref9 article-title: Learning to predict context-adaptive convolution for semantic segmentation publication-title: arxiv 2004 08222 – ident: ref20 doi: 10.1109/CVPR.2016.90 – ident: ref11 doi: 10.1109/TPAMI.2015.2389824 – ident: ref18 doi: 10.1109/TPAMI.2011.131 – ident: ref19 doi: 10.1016/j.patrec.2008.04.005 – ident: ref23 doi: 10.1109/ICCV.2017.533 – ident: ref25 doi: 10.1109/CVPR.2017.25 – ident: ref5 doi: 10.1109/TPAMI.2016.2644615 – ident: ref10 doi: 10.1109/ICCV.2015.178 – year: 2014 ident: ref13 article-title: Semantic image segmentation with deep convolutional nets and fully connected CRFs publication-title: arXiv 1412 7062 – year: 2017 ident: ref7 article-title: Rethinking atrous convolution for semantic image segmentation publication-title: arXiv 1706 05587 – ident: ref14 doi: 10.1109/TPAMI.2017.2699184 – ident: ref37 doi: 10.1103/RevModPhys.54.235 – ident: ref30 doi: 10.1109/CVPR.2017.660 – start-page: 549 year: 2018 ident: ref34 article-title: Self-erasing network for integral object attention publication-title: Proc Adv Neural Inf Process Syst – ident: ref26 doi: 10.1109/CVPR.2017.243 – ident: ref27 doi: 10.1109/CVPR.2018.00256 – ident: ref21 doi: 10.1007/978-3-030-01249-6_34 – ident: ref2 doi: 10.1109/CVPR.2019.00017 – ident: ref32 doi: 10.1109/TPAMI.2018.2852750 – ident: ref1 doi: 10.1109/TGRS.2018.2815613 – ident: ref17 doi: 10.1007/978-3-642-33715-4_54 – ident: ref24 doi: 10.1109/TIM.2020.3004902 – ident: ref12 doi: 10.1109/TPAMI.2017.2708714 – ident: ref38 doi: 10.1007/s11263-007-0090-8 – ident: ref6 doi: 10.1109/CVPR.2017.549 – ident: ref36 doi: 10.1109/ICCV.2019.00216 – ident: ref8 doi: 10.1109/ICCV.2015.179 – ident: ref39 doi: 10.1109/CVPR.2013.79 – ident: ref33 doi: 10.1109/TMM.2015.2477035
SSID	ssj0014847
Score	2.593058
Snippet	When dealing with semantic segmentation, how to locate the object boundary information more accurately is a key problem to distinguish different objects...
SourceID	proquest crossref ieee
SourceType	Aggregation Database Enrichment Source Index Database Publisher
StartPage	1926
SubjectTerms	boundary location Coders Computer vision Conditional random fields Convolution Decoding encoder-decoder Encoders-Decoders Feature extraction fully convolution network Image segmentation Multilayers Semantic segmentation Semantics Task analysis
Title	Encoder-Decoder With Cascaded CRFs for Semantic Segmentation
URI	https://ieeexplore.ieee.org/document/9165236 https://www.proquest.com/docview/2522214703
Volume	31
WOSCitedRecordID	wos000647394100020&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVIEE databaseName: IEEE/IET Electronic Library customDbUrl: eissn: 1558-2205 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0014847 issn: 1051-8215 databaseCode: RIE dateStart: 19910101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT8JAEJ4A8aAHX2hE0fTgTSultN3dxIupEA-GGEHk1uyrSiLF0OLvd3YpaKIx8dQ97CbNN9ud7-vOA-Cc8kCpkBOXyyhyA-ppl3FUKUxHkqTIv5ltBzS6J_0-HY_ZQwUu17kwWmsbfKavzNDe5auZXJhfZS2kMqiboipUCSHLXK31jUFAbTMxpAttl6IfWyXIeKw1jAejIUpBHxUqej9qKyJ-OSHbVeXHUWz9S2_nf2-2C9slj3Rulobfg4rO9mHrW3XBOlx3M5OwPndvtX06z5Pi1Yl5bkLilRM_9nIHKasz0FOEdyJx8DItU5GyA3jqdYfxnVs2S3Clz8LCVZ2IhxFV6F0UkQFCLygRUiD_QI5AtSCSC8ajMPW5CgTKHh6mnAulhcfx2OocQi2bZfoIHOkTJVSgGUHtp0KK9my3cR6XnZTQ1GtAe4VeIstK4qahxVtiFYXHEot4YhBPSsQbcLFe876so_Hn7LrBeD2zhLcBzZWRkvJTyxMfGaRptuR1jn9fdQKbvglEsVGKTagV84U-hQ35UUzy-ZndRZ88dcP2
linkProvider	IEEE
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JS8NAFH7UKqgHd7FaNQdvGpukSWYGvEhtqViL2LrcwmzRgk2li7_fN9O0CorgKXOYgfC9ybzvy7wF4ITyUKmIE5fLOHZD6mmXcVQpTMeSpMi_mW0H9Ngi7TZ9fmZ3BTib58JorW3wmT43Q3uXrwZyYn6VVZDKoG6KF2AxCsPAn2Zrze8MQmrbiSFh8F2KnmyWIuOxSrfWeeyiGAxQo6L_o7Ym4pcbsn1VfhzG1sM01v_3bhuwljNJ53Jq-k0o6GwLVr_VF9yGi3pmUtaH7pW2T-epN351anxkguKVU7tvjBwkrU5H9xHgnsTBSz9PRsp24KFR79aabt4uwZUBi8auqsY8iqlC_6KIDBF8QYmQAhkIsgSqBZFcMB5HacBVKFD48CjlXCgtPI4HV3UXitkg03vgyIAooULNCKo_FVG0qO_jPC6rKaGpVwJ_hl4i81ripqXFW2I1hccSi3hiEE9yxEtwOl_zPq2k8efsbYPxfGYObwnKMyMl-cc2SgLkkKbdklfd_33VMSw3u7etpHXdvjmAlcCEpdiYxTIUx8OJPoQl-THujYZHdkd9Apojxz0
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=Encoder-Decoder+With+Cascaded+CRFs+for+Semantic+Segmentation&rft.jtitle=IEEE+transactions+on+circuits+and+systems+for+video+technology&rft.au=Ji%2C+Jian&rft.au=Shi%2C+Rui&rft.au=Li%2C+Sitong&rft.au=Chen%2C+Peng&rft.date=2021-05-01&rft.issn=1051-8215&rft.eissn=1558-2205&rft.volume=31&rft.issue=5&rft.spage=1926&rft.epage=1938&rft_id=info:doi/10.1109%2FTCSVT.2020.3015866&rft.externalDBID=n%2Fa&rft.externalDocID=10_1109_TCSVT_2020_3015866
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1051-8215&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1051-8215&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1051-8215&client=summon