Robust workflow recognition using holistic features and outlier-tolerant fused hidden markov models

Voulodimos, A; Grabner, H; Kosmopoulos, D; Van Gool, L; Varvarigou, T

dc.contributor.author	Voulodimos, A	en
dc.contributor.author	Grabner, H	en
dc.contributor.author	Kosmopoulos, D	en
dc.contributor.author	Van Gool, L	en
dc.contributor.author	Varvarigou, T	en
dc.date.accessioned	2014-03-01T02:46:58Z
dc.date.available	2014-03-01T02:46:58Z
dc.date.issued	2010	en
dc.identifier.issn	03029743	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32961
dc.subject	Model Complexity	en
dc.subject	Target Recognition	en
dc.subject	Hidden Markov Model	en
dc.subject	Observation Likelihood	en
dc.subject.other	Complex environments	en
dc.subject.other	Descriptors	en
dc.subject.other	Different process	en
dc.subject.other	Heavy occlusion	en
dc.subject.other	Illumination changes	en
dc.subject.other	Limited visibility	en
dc.subject.other	Model complexes	en
dc.subject.other	Multiple cameras	en
dc.subject.other	Multivariate Student	en
dc.subject.other	Real world environments	en
dc.subject.other	Target recognition	en
dc.subject.other	Visual behavior	en
dc.subject.other	Work-flows	en
dc.subject.other	Feature extraction	en
dc.subject.other	Management	en
dc.subject.other	Neural networks	en
dc.subject.other	Hidden Markov models	en
dc.title	Robust workflow recognition using holistic features and outlier-tolerant fused hidden markov models	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1007/978-3-642-15819-3_71	en
heal.identifier.secondary	http://dx.doi.org/10.1007/978-3-642-15819-3_71	en
heal.publicationDate	2010	en
heal.abstract	Monitoring real world environments such as industrial scenes is a challenging task due to heavy occlusions, resemblance of different processes, frequent illumination changes, etc. We propose a robust framework for recognizing workflows in such complex environments, boasting a threefold contribution: Firstly, we employ a novel holistic scene descriptor to efficiently and robustly model complex scenes, thus bypassing the very challenging tasks of target recognition and tracking. Secondly, we handle the problem of limited visibility and occlusions by exploiting redundancies through the use of merged information from multiple cameras. Finally, we use the multivariate Student-t distribution as the observation likelihood of the employed Hidden Markov Models, in order to further enhance robustness. We evaluate the performance of the examined approaches under real-life visual behavior understanding scenarios and we compare and discuss the obtained results. © 2010 Springer-Verlag Berlin Heidelberg.	en
heal.journalName	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)	en
dc.identifier.doi	10.1007/978-3-642-15819-3_71	en
dc.identifier.volume	6352 LNCS	en
dc.identifier.issue	PART 1	en
dc.identifier.spage	551	en
dc.identifier.epage	560	en