Robust visual behavior recognition

Kosmopoulos, D; Chatzis, SP

dc.contributor.author	Kosmopoulos, D	en
dc.contributor.author	Chatzis, SP	en
dc.date.accessioned	2014-03-01T01:34:27Z
dc.date.available	2014-03-01T01:34:27Z
dc.date.issued	2010	en
dc.identifier.issn	1053-5888	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/20740
dc.subject.classification	Engineering, Electrical & Electronic	en
dc.subject.other	Activity recognition	en
dc.subject.other	Model training	en
dc.subject.other	Multiple cameras	en
dc.subject.other	Multiple source	en
dc.subject.other	Pixel level	en
dc.subject.other	Real time	en
dc.subject.other	Recognition rates	en
dc.subject.other	Semi-supervised learning	en
dc.subject.other	Sequential data	en
dc.subject.other	Visual behavior	en
dc.subject.other	Behavioral research	en
dc.title	Robust visual behavior recognition	en
heal.type	journalArticle	en
heal.identifier.primary	10.1109/MSP.2010.937392	en
heal.identifier.secondary	5562658	en
heal.identifier.secondary	http://dx.doi.org/10.1109/MSP.2010.937392	en
heal.language	English	en
heal.publicationDate	2010	en
heal.abstract	In this article, we propose a novel framework for robust visual behavior understanding, capable of achieving high recognition rates in demanding real-life environments and in almost real time. Our approach is based on the utilization of holistic visual behavior understanding methods, which perform modeling directly at the pixel level. This way, we eliminate the world representation layer that can be a significant source of errors for the modeling algorithms. Our proposed system is based on the utilization of information from multiple cameras, aiming to alleviate the effects of occlusions and other similar artifacts, which are rather common in real-life installations. To effectively exploit the acquired information for the purpose of real-time activity recognition, appropriate methodologies for modeling of sequential data stemming from multiple sources are examined. Moreover, we explore the efficacy of the additional application of semisupervised learning methodologies, in an effort to reduce the cost of model training in a completely supervised fashion. The performance of the examined approaches is thoroughly evaluated under real-life visual behavior understanding scenarios, and the obtained results are compared and discussed. © 2010 IEEE.	en
heal.publisher	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC	en
heal.journalName	IEEE Signal Processing Magazine	en
dc.identifier.doi	10.1109/MSP.2010.937392	en
dc.identifier.isi	ISI:000283242900006	en
dc.identifier.volume	27	en
dc.identifier.issue	5	en
dc.identifier.spage	34	en
dc.identifier.epage	45	en