Double Additive Margin Softmax Loss for Face Recognition
Learning large-margin face features whose intra-class variance is small and inter-class diversity is one of important challenges in feature learning applying Deep Convolutional Neural Networks (DCNNs) for face recognition. Recently, an appealing line of research is to incorporate an angular margin i...
Saved in:
Published in | Applied sciences Vol. 10; no. 1; p. 60 |
---|---|
Main Authors | , , , |
Format | Journal Article |
Language | English |
Published |
Basel
MDPI AG
01.01.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Learning large-margin face features whose intra-class variance is small and inter-class diversity is one of important challenges in feature learning applying Deep Convolutional Neural Networks (DCNNs) for face recognition. Recently, an appealing line of research is to incorporate an angular margin in the original softmax loss functions for obtaining discriminative deep features during the training of DCNNs. In this paper we propose a novel loss function, termed as double additive margin Softmax loss (DAM-Softmax). The presented loss has a clearer geometrical explanation and can obtain highly discriminative features for face recognition. Extensive experimental evaluation of several recent state-of-the-art softmax loss functions are conducted on the relevant face recognition benchmarks, CASIA-Webface, LFW, CALFW, CPLFW, and CFP-FP. We show that the proposed loss function consistently outperforms the state-of-the-art. |
---|---|
AbstractList | Learning large-margin face features whose intra-class variance is small and inter-class diversity is one of important challenges in feature learning applying Deep Convolutional Neural Networks (DCNNs) for face recognition. Recently, an appealing line of research is to incorporate an angular margin in the original softmax loss functions for obtaining discriminative deep features during the training of DCNNs. In this paper we propose a novel loss function, termed as double additive margin Softmax loss (DAM-Softmax). The presented loss has a clearer geometrical explanation and can obtain highly discriminative features for face recognition. Extensive experimental evaluation of several recent state-of-the-art softmax loss functions are conducted on the relevant face recognition benchmarks, CASIA-Webface, LFW, CALFW, CPLFW, and CFP-FP. We show that the proposed loss function consistently outperforms the state-of-the-art. |
Author | Hou, Xielian Chen, Caikou Han, Guojiang Zhou, Shengwei |
Author_xml | – sequence: 1 givenname: Shengwei surname: Zhou fullname: Zhou, Shengwei – sequence: 2 givenname: Caikou orcidid: 0000-0002-6941-4338 surname: Chen fullname: Chen, Caikou – sequence: 3 givenname: Guojiang surname: Han fullname: Han, Guojiang – sequence: 4 givenname: Xielian surname: Hou fullname: Hou, Xielian |
BookMark | eNpNkEtLAzEUhYNUsNau_AMBlzJ685gksyzVqlARfKyHvKZMaZMxMyP6743URS8X7l18nHM452gSYvAIXRK4YayCW911BCCvgBM0pSBFwTiRk6P_DM37fgt5KsIUgSlSd3E0O48XzrVD--Xxs06bNuC32Ax7_Y3Xse9xExNeaevxq7dxEzIYwwU6bfSu9_P_O0Mfq_v35WOxfnl4Wi7WhaWVGgoiK6EsE95oQUrdGKgM4SUVOZN1CjiV1DhoOAVfOSM41Z47or0lWknp2AxdHXS7FD9H3w_1No4pZMualoypUkkuM3V9oGzKgZNv6i61e51-agL1Xzv1UTvsF_0sVy4 |
CitedBy_id | crossref_primary_10_3390_app10134453 |
Cites_doi | 10.1109/CVPR.2017.243 10.1109/LSP.2018.2822810 10.1109/WACV.2016.7477558 10.1109/CVPR.2019.00482 10.1007/978-3-319-46478-7_31 10.1145/3123266.3123359 10.1109/CVPR.2016.90 10.1109/CVPR.2015.7298594 10.1109/CVPR.2015.7298682 10.1007/978-3-7908-2604-3_16 10.1109/CVPR.2017.713 10.1109/FG.2015.7284836 10.1007/s11263-015-0816-y |
ContentType | Journal Article |
Copyright | 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
Copyright_xml | – notice: 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
DBID | AAYXX CITATION ABUWG AFKRA AZQEC BENPR CCPQU DWQXO PIMPY PQEST PQQKQ PQUKI PRINS |
DOI | 10.3390/app10010060 |
DatabaseName | CrossRef ProQuest Central (Alumni) ProQuest Central UK/Ireland ProQuest Central Essentials AUTh Library subscriptions: ProQuest Central ProQuest One Community College ProQuest Central Publicly Available Content Database ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China |
DatabaseTitle | CrossRef Publicly Available Content Database ProQuest Central ProQuest One Academic UKI Edition ProQuest Central Essentials ProQuest Central Korea ProQuest One Academic Eastern Edition ProQuest Central (Alumni Edition) ProQuest One Community College ProQuest One Academic ProQuest Central China |
DatabaseTitleList | Publicly Available Content Database CrossRef |
Database_xml | – sequence: 1 dbid: BENPR name: AUTh Library subscriptions: ProQuest Central url: https://www.proquest.com/central sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering Sciences (General) |
EISSN | 2076-3417 |
ExternalDocumentID | 10_3390_app10010060 |
GroupedDBID | .4S 5VS 7XC 8CJ 8FE 8FG 8FH AADQD AAFWJ AAYXX ABJCF ADBBV AFKRA AFPKN AFZYC ALMA_UNASSIGNED_HOLDINGS APEBS ARAPS ARCSS ATCPS BBNVY BCNDV BENPR BHPHI BKSAR CCPQU CITATION CZ9 D1I D1J D1K GROUPED_DOAJ HCIFZ IAO K6- K6V K7- KB. KC. KQ8 L6V LK5 LK8 M0K M7P M7R M7S MODMG M~E OK1 P62 PATMY PCBAR PDBOC PIMPY PROAC PYCSY RIG TUS 2XV ABUWG AZQEC DWQXO ITC PQEST PQQKQ PQUKI PRINS |
ID | FETCH-LOGICAL-c298t-17968c36eba615afb09b14526207cd804272bd0f420e9db642ae4d1aec1a877d3 |
IEDL.DBID | BENPR |
ISSN | 2076-3417 |
IngestDate | Thu Oct 10 18:12:01 EDT 2024 Fri Aug 23 04:54:35 EDT 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 1 |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c298t-17968c36eba615afb09b14526207cd804272bd0f420e9db642ae4d1aec1a877d3 |
ORCID | 0000-0002-6941-4338 |
OpenAccessLink | https://www.proquest.com/docview/2533858747?pq-origsite=%requestingapplication% |
PQID | 2533858747 |
PQPubID | 2032433 |
ParticipantIDs | proquest_journals_2533858747 crossref_primary_10_3390_app10010060 |
PublicationCentury | 2000 |
PublicationDate | 2020-01-01 |
PublicationDateYYYYMMDD | 2020-01-01 |
PublicationDate_xml | – month: 01 year: 2020 text: 2020-01-01 day: 01 |
PublicationDecade | 2020 |
PublicationPlace | Basel |
PublicationPlace_xml | – name: Basel |
PublicationTitle | Applied sciences |
PublicationYear | 2020 |
Publisher | MDPI AG |
Publisher_xml | – name: MDPI AG |
References | Russakovsky (ref_8) 2015; 115 ref_14 ref_13 ref_24 ref_12 ref_23 ref_11 ref_22 ref_10 ref_21 ref_20 ref_1 ref_3 ref_2 ref_19 ref_18 ref_17 ref_16 ref_9 Wang (ref_15) 2018; 25 ref_5 ref_4 ref_7 ref_6 |
References_xml | – ident: ref_7 – ident: ref_6 doi: 10.1109/CVPR.2017.243 – volume: 25 start-page: 926 year: 2018 ident: ref_15 article-title: Additive margin softmax for face verification publication-title: IEEE Signal Process. Lett. doi: 10.1109/LSP.2018.2822810 contributor: fullname: Wang – ident: ref_18 doi: 10.1109/WACV.2016.7477558 – ident: ref_3 – ident: ref_22 doi: 10.1109/CVPR.2019.00482 – ident: ref_2 – ident: ref_10 doi: 10.1007/978-3-319-46478-7_31 – ident: ref_12 – ident: ref_24 doi: 10.1145/3123266.3123359 – ident: ref_11 – ident: ref_5 doi: 10.1109/CVPR.2016.90 – ident: ref_16 – ident: ref_4 doi: 10.1109/CVPR.2015.7298594 – ident: ref_13 – ident: ref_1 doi: 10.1109/CVPR.2015.7298682 – ident: ref_17 – ident: ref_9 doi: 10.1007/978-3-7908-2604-3_16 – ident: ref_14 doi: 10.1109/CVPR.2017.713 – ident: ref_19 – ident: ref_23 – ident: ref_20 doi: 10.1109/FG.2015.7284836 – ident: ref_21 – volume: 115 start-page: 211 year: 2015 ident: ref_8 article-title: ImageNet Large Scale Visual Recognition Challenge publication-title: Int. J. Comput. Vis. doi: 10.1007/s11263-015-0816-y contributor: fullname: Russakovsky |
SSID | ssj0000913810 |
Score | 2.1998534 |
Snippet | Learning large-margin face features whose intra-class variance is small and inter-class diversity is one of important challenges in feature learning applying... |
SourceID | proquest crossref |
SourceType | Aggregation Database |
StartPage | 60 |
SubjectTerms | Benchmarks Euclidean space Face recognition Feature recognition Learning Neural networks Pattern recognition |
Title | Double Additive Margin Softmax Loss for Face Recognition |
URI | https://www.proquest.com/docview/2533858747 |
Volume | 10 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1NT8JAEJ0IXPRgBDWiSPbAQQ-N3W1ptyeDhkqMEoOScGv2q4mJFhRM_PnulK3CxXMPm8x0Z9587HsAvURHTCkderGkgRf2pfG45LknIxFEORUqovjA-XEcjabh_aw_cw23pVurrGJiGaj1XGGP_IpZXML73KLf68WHh6pROF11Eho1aDAa4pi2cTMcP01-uyzIesmpv36YF9j6HufCSDuEPCTbqWg7EpfpJT2AfYcLyWDtyCbsmKIFextsgS1ounu4JBeOLPryELhFwPLNkIHW5RYQQeXa14I82_D6Lr7Jgz2ZWGBKUqEMmVTrQvPiCKbp8OV25Dk1BE-xhK-QRzTiKoiMFBaFiFz6iaQoEM78WGmOmhlMaj8PmW8SLW1dIUyoqTCKCh7HOjiGejEvzAkQHcbWP4IrxkwYC5lY0GrPwPupfSp0G3qVYbLFmvQis8UC2i_bsF8bOpXRMvfnL7M_P53-__kMdhnWrmU7owP11eeXObcJfiW7UOPpXdf5sluWyT_0WKeB |
link.rule.ids | 315,786,790,870,12792,21416,27955,27956,33406,33777,43633,43838,74390,74657 |
linkProvider | ProQuest |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV09T8MwED1BGYAB0QKiUMBDBxgiYsd1nAlViFCg7QCt1C3yVyQkSAstEj8fO3WgXZg9WLqz796dz-8BtBPNiFKaBrHEUUA70gRc8jyQTEQsx0Ix7D44D4asN6aPk87EN9zmfqyyiolloNZT5Xrk18TiEt7hFv3ezD4CpxrlXle9hMYmbNGIRe6c8_T-t8fiOC85Dpff8iJb3btXYUc65FhI1hPRehwuk0u6D3seFaLu0o112DBFA3ZXuAIbUPe3cI4uPVX01QFwi3_lm0FdrcsZIOR0a18L9GKD67v4Rn27M7KwFKVCGfRcDQtNi0MYp3ej217gtRACRRK-cCyijKuIGSksBhG5DBOJnTw4CWOluVPMIFKHOSWhSbS0VYUwVGNhFBY8jnV0BLViWphjQJrG1juCK0IMjYVMLGS1e7jbqUMsdBPalWGy2ZLyIrOlgrNftmK_JrQqo2X-3M-zPy-d_L98Adu90aCf9R-GT6ewQ1wVWzY2WlBbfH6ZM5vqF_K89OcPMnmnLQ |
linkToPdf | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1NS8NAEB20guhBbFWsVt1DD3oIzW7SZHOSosaqtYha6C3sV0DQtNoK_nx30o3Wi-c9LMwwM29mZ98DaCc6Ykrp0IslDbywK43HJc89GYkgyqlQEcUPzvfDqD8Kb8fdsdt_mrm1yionlolaTxTOyDvM4hLe5Rb9dnK3FvFwmZ5P3z1UkMKXViensQprCLJRxoGn1z_zFuS_5NRffNEL7Dm-ECMBETKS_C1Kf3NyWWjSbdhyCJH0Fi6tw4opGrC5xBvYgLqLyBk5dbTRZzvALRaWr4b0tC73gQhq2L4U5Mkm2jfxRQb2ZmIhKkmFMuSxWhyaFLswSq-eL_qe00XwFEv4HBlFI66CyEhh8YjIpZ9IilLhzI-V5qiewaT285D5JtHSdhjChJoKo6jgcayDPagVk8LsA9FhbD0luGLMhLGQiYWv9g6MVO1ToZvQrgyTTRf0F5ltG9B-2ZL9mtCqjJa5GJhlvx47-P_4BNatK7PBzfDuEDYYNrTljKMFtfnHpzmyVX8uj0t3fgPO0qti |
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=Double+Additive+Margin+Softmax+Loss+for+Face+Recognition&rft.jtitle=Applied+sciences&rft.au=Zhou%2C+Shengwei&rft.au=Chen%2C+Caikou&rft.au=Han%2C+Guojiang&rft.au=Hou%2C+Xielian&rft.date=2020-01-01&rft.issn=2076-3417&rft.eissn=2076-3417&rft.volume=10&rft.issue=1&rft.spage=60&rft_id=info:doi/10.3390%2Fapp10010060&rft.externalDBID=n%2Fa&rft.externalDocID=10_3390_app10010060 |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2076-3417&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2076-3417&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2076-3417&client=summon |