A Kinect-based Wearable Face Recognition System To Aid Visually Impaired Users

dc.creatorNeto
dc.creatorLaurindo Britto; Grijalva
dc.creatorFelipe; Margareth Lima Maike
dc.creatorVanessa Regina; Martini
dc.creatorLuiz Cesar; Florencio
dc.creatorDinei; Calani Baranauskas
dc.creatorMaria Cecilia; Rocha
dc.creatorAnderson; Goldenstein
dc.creatorSiome
dc.date2017
dc.datefev
dc.date2017-11-13T13:44:54Z
dc.date2017-11-13T13:44:54Z
dc.date.accessioned2018-03-29T05:59:34Z
dc.date.available2018-03-29T05:59:34Z
dc.identifierIeee Transactions On Human-machine Systems. Ieee-inst Electrical Electronics Engineers Inc, v. 47, p. 52 - 64, 2017.
dc.identifier2168-2291
dc.identifier2168-2305
dc.identifierWOS:000396400300006
dc.identifier10.1109/THMS.2016.2604367
dc.identifierhttp://ieeexplore.ieee.org/document/7571103/
dc.identifierhttp://repositorio.unicamp.br/jspui/handle/REPOSIP/328886
dc.identifier.urihttp://repositorioslatinoamericanos.uchile.cl/handle/2250/1365911
dc.descriptionFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.descriptionFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.descriptionCoordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
dc.descriptionCoordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
dc.descriptionIn this paper, we introduce a real-time face recognition (and announcement) system targeted at aiding the blind and low-vision people. The system uses a Microsoft Kinect sensor as a wearable device, performs face detection, and uses temporal coherence along with a simple biometric procedure to generate a sound associated with the identified person, virtualized at his/her estimated 3-D location. Our approach uses a variation of the K-nearest neighbors algorithm over histogram of oriented gradient descriptors dimensionally reduced by principal component analysis. The results show that our approach, on average, outperforms traditional face recognition methods while requiring much less computational resources (memory, processing power, and battery life) when compared with existing techniques in the literature, deeming it suitable for the wearable hardware constraints. We also show the performance of the system in the dark, using depth-only information acquired with Kinect's infrared camera. The validation uses a new dataset available for download, with 600 videos of 30 people, containing variation of illumination, background, and movement patterns. Experiments with existing datasets in the literature are also considered. Finally, we conducted user experience evaluations on both blindfolded and visually impaired users, showing encouraging results.
dc.description47
dc.description1
dc.description52
dc.description64
dc.descriptionMicrosoft-Sao Paulo Research Foundation (FAPESP) [2012/50468-6]
dc.descriptionUnicamp Institutional Review Board [CAAE 15641313.7.0000.5404, CAAE 31818014.0.0000.5404]
dc.descriptionCNPq [141254/2014-9, 308618/2014-9, 304352/2012-8, 308882/2013-0, 304472/2015-8, 477662/2013-7]
dc.descriptionFAPESP [2014/14630-9, 2013/21349-1, 2015/19222-9]
dc.descriptionDejaVu project [2015/19222-9]
dc.descriptionCAPES [01-P-04554/2013]
dc.descriptionCAPES DeepEyes project
dc.descriptionFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.descriptionConselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)
dc.descriptionFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
dc.descriptionCoordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
dc.descriptionCoordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)
dc.languageEnglish
dc.publisherIEEE-Inst Electrical Electronics Engineers Inc
dc.publisherPiscataway
dc.relationIEEE Transactions on Human-machine Systems
dc.rightsfechado
dc.sourceWOS
dc.subjectAccessibility
dc.subjectAssistive Technology
dc.subjectFace Recognition
dc.subjectMicrosoft Kinect
dc.subjectWearable Device
dc.subjectWearable System
dc.titleA Kinect-based Wearable Face Recognition System To Aid Visually Impaired Users
dc.typeArtículos de revistas


Este ítem pertenece a la siguiente institución