Compound document analysis by fusing evidence across media

Nikolopoulos, S; Lakka, C; Kompatsiaris, I; Varytimidis, C; Rapantzikos, K; Avrithis, Y

dc.contributor.author	Nikolopoulos, S	en
dc.contributor.author	Lakka, C	en
dc.contributor.author	Kompatsiaris, I	en
dc.contributor.author	Varytimidis, C	en
dc.contributor.author	Rapantzikos, K	en
dc.contributor.author	Avrithis, Y	en
dc.date.accessioned	2014-03-01T02:46:04Z
dc.date.available	2014-03-01T02:46:04Z
dc.date.issued	2009	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32521
dc.subject	Document Analysis	en
dc.subject	Heterogeneous Media	en
dc.subject	Performance Improvement	en
dc.subject	Probabilistic Inference	en
dc.subject	Semantic Interpretation	en
dc.subject	bayesian network	en
dc.subject.other	Compound document	en
dc.subject.other	Cross-media	en
dc.subject.other	Cross-media relations	en
dc.subject.other	Fusion mechanism	en
dc.subject.other	Heterogeneous media	en
dc.subject.other	Performance improvements	en
dc.subject.other	Probabilistic inference	en
dc.subject.other	Semantic interpretation	en
dc.subject.other	Bayesian networks	en
dc.subject.other	Indexing (of information)	en
dc.subject.other	Inference engines	en
dc.subject.other	Multimedia systems	en
dc.subject.other	Security of data	en
dc.subject.other	Program documentation	en
dc.title	Compound document analysis by fusing evidence across media	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/CBMI.2009.35	en
heal.identifier.secondary	http://dx.doi.org/10.1109/CBMI.2009.35	en
heal.identifier.secondary	5137837	en
heal.publicationDate	2009	en
heal.abstract	In this paper a cross media analysis scheme for the semantic interpretation of compound documents is presented. It is essentially a late-fusion mechanism that operates on top of single-media extractors output and it's main novelty relies on using the evidence extracted from heterogeneous media sources to perform probabilistic inference on a bayesian network that incorporates knowledge about the domain. Experiments performed on a set of 54 compound documents showed that the proposed scheme is able to exploit the existing cross media relations and achieve performance improvements. © 2009 IEEE.	en
heal.journalName	7th International Workshop on Content-Based Multimedia Indexing, CBMI 2009	en
dc.identifier.doi	10.1109/CBMI.2009.35	en
dc.identifier.spage	175	en
dc.identifier.epage	180	en