Please use this identifier to cite or link to this item: http://hdl.handle.net/10553/70579
DC FieldValueLanguage
dc.contributor.authorMarras, Mirkoen_US
dc.contributor.authorMarín-Reyes, Pedro A.en_US
dc.contributor.authorLorenzo-Navarro, Javieren_US
dc.contributor.authorCastrillón-Santana, Modestoen_US
dc.contributor.authorFenu, Giannien_US
dc.date.accessioned2020-02-29T06:04:01Z-
dc.date.available2020-02-29T06:04:01Z-
dc.date.issued2020en_US
dc.identifier.isbn978-3-030-40013-2en_US
dc.identifier.issn0302-9743en_US
dc.identifier.otherScopus-
dc.identifier.urihttp://hdl.handle.net/10553/70579-
dc.description.abstractFrom border controls to personal devices, from online exam proctoring to human-robot interaction, biometric technologies are empowering individuals and organizations with convenient and secure authentication and identification services. However, most biometric systems leverage only a single modality, and may face challenges related to acquisition distance, environmental conditions, data quality, and computational resources. Combining evidence from multiple sources at a certain level (e.g., sensor, feature, score, or decision) of the recognition pipeline may mitigate some limitations of the common uni-biometric systems. Such a fusion has been rarely investigated at intermediate level, i.e., when uni-biometric model parameters are jointly optimized during training. In this chapter, we propose a multi-biometric model training strategy that digests face and voice traits in parallel, and we explore how it helps to improve recognition performance in re-identification and verification scenarios. To this end, we design a neural architecture for jointly embedding face and voice data, and we experiment with several training losses and audio-visual datasets. The idea is to exploit the relation between voice characteristics and facial morphology, so that face and voice uni-biometric models help each other to recognize people when trained jointly. Extensive experiments on four real-world datasets show that the biometric feature representation of a uni-biometric model jointly trained performs better than the one computed by the same uni-biometric model trained alone. Moreover, the recognition results are further improved by embedding face and voice data into a single shared representation of the two modalities. The proposed fusion strategy generalizes well on unseen and unheard users, and should be considered as a feasible solution that improves model performance. We expect that this chapter will support the biometric community to shape the research on deep audio-visual fusion in real-world contexts.en_US
dc.languageengen_US
dc.publisherSpringeren_US
dc.relationIdentificación Automática de Oradores en Sesiones Parlamentarias Usando Características Audiovisuales.en_US
dc.relation.ispartofLecture Notes in Computer Scienceen_US
dc.sourcePattern Recognition Applications and Methods. ICPRAM 2019. Lecture Notes in Computer Science, v. 11996, p. 136-157en_US
dc.subject120304 Inteligencia artificialen_US
dc.subject.otherAudio-visual learningen_US
dc.subject.otherCross-modal biometricsen_US
dc.subject.otherDeep biometric fusionen_US
dc.subject.otherMulti-biometric systemen_US
dc.subject.otherRe-identificationen_US
dc.subject.otherVerificationen_US
dc.titleDeep multi-biometric fusion for audio-visual user re-identification and verificationen_US
dc.typeinfo:eu-repo/semantics/bookParten_US
dc.typeBook parten_US
dc.relation.conference8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019
dc.identifier.doi10.1007/978-3-030-40014-9_7en_US
dc.identifier.scopus85079549512-
dc.contributor.authorscopusid9233842500-
dc.contributor.authorscopusid57191274555-
dc.contributor.authorscopusid15042453800-
dc.contributor.authorscopusid57198776493-
dc.contributor.authorscopusid24469552000-
dc.description.lastpage157en_US
dc.description.firstpage136en_US
dc.relation.volume11996en_US
dc.investigacionIngeniería y Arquitecturaen_US
dc.type2Capítulo de libroen_US
dc.identifier.eisbn978-3-030-40014-9-
dc.utils.revisionen_US
dc.identifier.supplement0302-9743-
dc.identifier.supplement0302-9743-
dc.identifier.supplement0302-9743-
dc.identifier.supplement0302-9743-
dc.identifier.conferenceidevents121650-
dc.identifier.ulpgcen_US
dc.identifier.ulpgcen_US
dc.identifier.ulpgcen_US
dc.identifier.ulpgcen_US
dc.contributor.buulpgcBU-INFen_US
dc.contributor.buulpgcBU-INFen_US
dc.contributor.buulpgcBU-INFen_US
dc.contributor.buulpgcBU-INFen_US
dc.description.sjr0,249
dc.description.sjrqQ3
dc.description.spiqQ1
item.grantfulltextnone-
item.fulltextSin texto completo-
crisitem.author.deptGIR SIANI: Inteligencia Artificial, Robótica y Oceanografía Computacional-
crisitem.author.deptIU Sistemas Inteligentes y Aplicaciones Numéricas-
crisitem.author.deptDepartamento de Informática y Sistemas-
crisitem.author.deptGIR SIANI: Inteligencia Artificial, Robótica y Oceanografía Computacional-
crisitem.author.deptIU Sistemas Inteligentes y Aplicaciones Numéricas-
crisitem.author.deptDepartamento de Informática y Sistemas-
crisitem.author.orcid0000-0002-2834-2067-
crisitem.author.orcid0000-0002-8673-2725-
crisitem.author.parentorgIU Sistemas Inteligentes y Aplicaciones Numéricas-
crisitem.author.parentorgIU Sistemas Inteligentes y Aplicaciones Numéricas-
crisitem.author.fullNameMarín Reyes, Pedro Antonio-
crisitem.author.fullNameLorenzo Navarro, José Javier-
crisitem.author.fullNameCastrillón Santana, Modesto Fernando-
crisitem.project.principalinvestigatorCastrillón Santana, Modesto Fernando-
crisitem.event.eventsstartdate19-02-2019-
crisitem.event.eventsenddate21-02-2019-
Appears in Collections:Capítulo de libro
Show simple item record

SCOPUSTM   
Citations

7
checked on Nov 27, 2022

Page view(s)

131
checked on Oct 1, 2022

Google ScholarTM

Check

Altmetric


Share



Export metadata



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