Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/320184
Full metadata record
DC FieldValueLanguage
dc.contributor.CRUESPUNIVERSIDADE ESTADUAL DE CAMPINASpt_BR
dc.identifier.isbn1095-9076pt
dc.contributor.authorunicampRocha, Anderson de Rezendept_BR
dc.contributor.authorunicampChiachia, Giovanipt_BR
dc.typeArtigopt_BR
dc.titleRate-energy-accuracy optimization of convolutional architectures for face recognitionpt_BR
dc.contributor.authorBondi, L.pt_BR
dc.contributor.authorBaroffio, L.pt_BR
dc.contributor.authorCesana, M.pt_BR
dc.contributor.authorTagliasacchi, M.pt_BR
dc.contributor.authorChiachia, G.pt_BR
dc.contributor.authorRocha, A.pt_BR
unicamp.author.emailluca.bondi@polimi.it; luca.baroffio@polimi.it; matteo.cesana@polimi.it; marco.taglisacchi@polimi.it; chiachia@ic.unicamp.br; rocha@ic.unicamp.brpt_BR
dc.subjectRedes neurais convolucionaispt_BR
dc.subjectAprendizado profundopt_BR
dc.subjectRedes neurais (Computação)pt_BR
dc.subjectReconhecimento facial (Computação)pt_BR
dc.subject.otherlanguageConvolutional neural networkspt_BR
dc.subject.otherlanguageDeep learningpt_BR
dc.subject.otherlanguageNeural networks (Computer science)pt_BR
dc.subject.otherlanguageHuman face recognition (Computer science)pt_BR
dc.description.abstractFace recognition systems based on Convolutional Neural Networks (CNNs) or convolutional architectures currently represent the state of the art, achieving an accuracy comparable to that of humans. Nonetheless, there are two issues that might hinder their adoption on distributed battery-operated devices (e.g., visual sensor nodes, smartphones, and wearable devices). First, convolutional architectures are usually computationally demanding, especially when the depth of the network is increased to maximize accuracy. Second, transmitting the output features produced by a CNN might require a bitrate higher than the one needed for coding the input image. Therefore, in this paper we address the problem of optimizing the energy-rate-accuracy characteristics of a convolutional architecture for face recognition. We carefully profile a CNN implementation on a Raspberry Pi device and optimize the structure of the neural network, achieving a 17-fold speedup without significantly affecting recognition accuracy. Moreover, we propose a coding architecture custom-tailored to features extracted by such model. (C) 2015 Elsevier Inc. All rights reserved.en
dc.description.abstractFace recognition systems based on Convolutional Neural Networks (CNNs) or convolutional architectures currently represent the state of the art, achieving an accuracy comparable to that of humans. Nonetheless, there are two issues that might hinder their apt_BR
dc.relation.ispartofJournal of visual communication and image representationpt_BR
dc.publisher.cityOxfordpt_BR
dc.publisher.countryReino Unidopt_BR
dc.publisherElsevierpt_BR
dc.date.issued2016pt_BR
dc.date.monthofcirculationApr.pt_BR
dc.identifier.citationJournal Of Visual Communication And Image Representation. ACADEMIC PRESS INC ELSEVIER SCIENCE, n. 36, p. 142 - 148.pt_BR
dc.language.isoengpt_BR
dc.description.volume36pt_BR
dc.description.issuenumberpt_BR
dc.description.issuesupplementpt_BR
dc.description.issuepartpt_BR
dc.description.issuespecialpt_BR
dc.description.firstpage142pt_BR
dc.description.lastpage148pt_BR
dc.rightsfechadopt_BR
dc.rightsAbertopt_br
dc.sourceWOSpt_BR
dc.identifier.issn1047-3203pt_BR
dc.identifier.eissn1095-9076pt_BR
dc.identifier.wosidWOS:000371280200012pt_BR
dc.identifier.doi10.1016/j.jvcir.2015.12.015pt_BR
dc.identifier.urlhttps://www.sciencedirect.com/science/article/pii/S1047320315002552pt_BR
dc.description.sponsorshipCNPQ - CONSELHO NACIONAL DE DESENVOLVIMENTO CIENTÍFICO E TECNOLÓGICOpt_BR
dc.description.sponsorshipCAPES - COORDENAÇÃO DE APERFEIÇOAMENTO DE PESSOAL DE NÍVEL SUPERIORpt_BR
dc.description.sponsorshipFAPESP - FUNDAÇÃO DE AMPARO À PESQUISA DO ESTADO DE SÃO PAULOpt_BR
dc.description.sponsorship1Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)pt_BR
dc.description.sponsorship1Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)pt_BR
dc.description.sponsorship1Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)pt_BR
dc.description.sponsordocumentnumbersem informaçãopt_BR
dc.description.sponsordocumentnumber2013/11359-0pt_BR
dc.description.sponsordocumentnumbersem informaçãopt_BR
dc.date.available2016-12-06T18:30:59Z-
dc.date.accessioned2016-12-06T18:30:59Z-
dc.description.provenanceMade available in DSpace on 2016-12-06T18:30:59Z (GMT). No. of bitstreams: 1 000371280200012.pdf: 1435014 bytes, checksum: c4a0b031c8b599111700ee56ee1a1854 (MD5) Previous issue date: 2016 Bitstreams deleted on 2021-01-04T14:26:12Z: 000371280200012.pdf,. Added 1 bitstream(s) on 2021-01-04T14:27:15Z : No. of bitstreams: 1 000371280200012.pdf: 1506670 bytes, checksum: ade57683d617bb78dfc44594ae1e6333 (MD5)en
dc.identifier.urihttp://repositorio.unicamp.br/jspui/handle/REPOSIP/320184-
dc.description.conferencenomept_BR
dc.contributor.departmentDepartamento de Sistemas de Informaçãopt_BR
dc.contributor.departmentsem informaçãopt_BR
dc.contributor.unidadeInstituto de Computaçãopt_BR
dc.contributor.unidadeInstituto de Computaçãopt_BR
dc.subject.keywordConvolutional architecturespt_BR
dc.subject.keywordConvolutional neural networks (CNNs)pt_BR
dc.subject.keywordOptimizationpt_BR
dc.subject.keywordCodingpt_BR
dc.subject.keywordFace recognitionpt_BR
dc.subject.keywordAnalyze-then-Compress (ATC)pt_BR
dc.subject.keywordDeep learningpt_BR
dc.subject.keywordDeep neural networkspt_BR
dc.identifier.source000371280200012pt_BR
dc.creator.orcid0000-0002-4236-8217pt_BR
dc.creator.orcid0000-0002-9622-1734pt_BR
dc.type.formArtigo de pesquisapt_BR
Appears in Collections:IC - Artigos e Outros Documentos

Files in This Item:
File SizeFormat 
000371280200012.pdf1.47 MBAdobe PDFView/Open


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