Real-time head pose estimation using multi-task deep neural network

Cited 42 time in webofscience Cited 0 time in scopus
  • Hit : 708
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorAhn, Byungtaeko
dc.contributor.authorChoi, Dong-Geolko
dc.contributor.authorPark, Jaesikko
dc.contributor.authorKweon, In Soko
dc.date.accessioned2018-05-24T02:23:37Z-
dc.date.available2018-05-24T02:23:37Z-
dc.date.created2018-05-08-
dc.date.created2018-05-08-
dc.date.created2018-05-08-
dc.date.issued2018-05-
dc.identifier.citationROBOTICS AND AUTONOMOUS SYSTEMS, v.103, pp.1 - 12-
dc.identifier.issn0921-8890-
dc.identifier.urihttp://hdl.handle.net/10203/242240-
dc.description.abstractDriver inattention is one of the main causes of traffic accidents. To avoid such accidents, advanced driver assistance system that passively monitors the driver's activities is needed. In this paper, we present a novel method to estimate a head pose from a monocular camera. The proposed algorithm is based on multi-task learning deep neural network that uses a small grayscale image. The network jointly detects multi-view faces and estimates head pose even under poor environment conditions such as illumination change, vibration, large pose change, and occlusion. We also propose a multi-task learning method that does not bias on a specific task with different datasets. Moreover, in order to fertilize training dataset, we establish and release the RCVFace dataset that has accurate head poses. The proposed framework outperforms state-of-the-art approaches quantitatively and qualitatively with an average head pose mean error of less than 4 in real-time. The algorithm applies to driver monitoring system that is crucial for driver safety. (C) 2018 Published by Elsevier B.V.-
dc.languageEnglish-
dc.publisherELSEVIER SCIENCE BV-
dc.subjectFACE ALIGNMENT-
dc.subjectMODELS-
dc.subjectLOCALIZATION-
dc.subjectTRACKING-
dc.titleReal-time head pose estimation using multi-task deep neural network-
dc.typeArticle-
dc.identifier.wosid000430764100001-
dc.identifier.scopusid2-s2.0-85044858033-
dc.type.rimsART-
dc.citation.volume103-
dc.citation.beginningpage1-
dc.citation.endingpage12-
dc.citation.publicationnameROBOTICS AND AUTONOMOUS SYSTEMS-
dc.identifier.doi10.1016/j.robot.2018.01.005-
dc.contributor.localauthorKweon, In So-
dc.contributor.nonIdAuthorAhn, Byungtae-
dc.description.isOpenAccessN-
dc.type.journalArticleArticle-
dc.subject.keywordAuthorHead pose-
dc.subject.keywordAuthorAdvanced driver assistance system-
dc.subject.keywordAuthorDeep learning-
dc.subject.keywordAuthorConvolutional neural network-
dc.subject.keywordPlusFACE ALIGNMENT-
dc.subject.keywordPlusMODELS-
dc.subject.keywordPlusLOCALIZATION-
dc.subject.keywordPlusTRACKING-
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 42 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0