2D and 3D face localization for complex scenes

Karame, G; Stergiou, A; Katsarakis, N; Papageorgiou, P; Pnevmatikakis, A

dc.contributor.author	Karame, G	en
dc.contributor.author	Stergiou, A	en
dc.contributor.author	Katsarakis, N	en
dc.contributor.author	Papageorgiou, P	en
dc.contributor.author	Pnevmatikakis, A	en
dc.date.accessioned	2014-03-01T02:51:00Z
dc.date.available	2014-03-01T02:51:00Z
dc.date.issued	2007	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/35288
dc.subject.other	Cameras	en
dc.subject.other	Computer networks	en
dc.subject.other	Face recognition	en
dc.subject.other	Stochastic models	en
dc.subject.other	Stochastic programming	en
dc.subject.other	Three dimensional	en
dc.subject.other	(algorithmic) complexity	en
dc.subject.other	3D faces	en
dc.subject.other	3D scenes	en
dc.subject.other	camera view	en
dc.subject.other	Complex scenes	en
dc.subject.other	Existing systems	en
dc.subject.other	Face localization	en
dc.subject.other	Face tracking	en
dc.subject.other	far fields	en
dc.subject.other	Gaussian mixture model (GMM)	en
dc.subject.other	Multiple people	en
dc.subject.other	Security systems	en
dc.title	2D and 3D face localization for complex scenes	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/AVSS.2007.4425339	en
heal.identifier.secondary	http://dx.doi.org/10.1109/AVSS.2007.4425339	en
heal.identifier.secondary	4425339	en
heal.publicationDate	2007	en
heal.abstract	In this paper, we address face tracking of multiple people in complex 3D scenes, using multiple calibrated and synchronized far-field recordings. We localize faces in every camera view and associate them across the different views. To cope with the complexity of 2D face localization introduced by the multitude of people and unconstrained face poses, a combination of stochastic and deterministic trackers, detectors and a Gaussian Mixture Model for face validation are utilized. Then faces of the same person seen from the different cameras are associated by first finding all possible associations and then choosing the best option by means of a 3D stochastic tracker. The performance of the proposed system is evaluated and is found enhanced compared to existing systems. © 2007 IEEE.	en
heal.journalName	2007 IEEE Conference on Advanced Video and Signal Based Surveillance, AVSS 2007 Proceedings	en
dc.identifier.doi	10.1109/AVSS.2007.4425339	en
dc.identifier.spage	371	en
dc.identifier.epage	376	en