Stochastic framework for optimal key frame extraction from MPEG video databases

Avrithis, YS; Doulamis, AD; Doulamis, ND; Kollias, SD

dc.contributor.author	Avrithis, YS	en
dc.contributor.author	Doulamis, AD	en
dc.contributor.author	Doulamis, ND	en
dc.contributor.author	Kollias, SD	en
dc.date.accessioned	2014-03-01T01:14:21Z
dc.date.available	2014-03-01T01:14:21Z
dc.date.issued	1999	en
dc.identifier.issn	1077-3142	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/13015
dc.subject	Cross Correlation	en
dc.subject	Efficient Implementation	en
dc.subject	Feature Vector	en
dc.subject	Genetic Algorithm	en
dc.subject	Motion Segmentation	en
dc.subject	Mpeg Video	en
dc.subject	Spanning Tree	en
dc.subject	Temporal Variation	en
dc.subject.classification	Computer Science, Artificial Intelligence	en
dc.subject.classification	Engineering, Electrical & Electronic	en
dc.subject.other	Database systems	en
dc.subject.other	Genetic algorithms	en
dc.subject.other	Image analysis	en
dc.subject.other	Image compression	en
dc.subject.other	Image quality	en
dc.subject.other	Image segmentation	en
dc.subject.other	Standards	en
dc.subject.other	Trees (mathematics)	en
dc.subject.other	Motion picture experts group (MPEG) standards	en
dc.subject.other	Recursive shortest spanning tree (RSST) algorithms	en
dc.subject.other	Feature extraction	en
dc.title	Stochastic framework for optimal key frame extraction from MPEG video databases	en
heal.type	journalArticle	en
heal.identifier.primary	10.1006/cviu.1999.0761	en
heal.identifier.secondary	http://dx.doi.org/10.1006/cviu.1999.0761	en
heal.language	English	en
heal.publicationDate	1999	en
heal.abstract	A video content representation framework is proposed in this paper for extracting limited, but meaningful, information of video data, directly from the MPEG compressed domain. A hierarchical color and motion segmentation scheme is applied to each video shot, transforming the frame-based representation to a feature-based one. The scheme is based on a multiresolution implementation of the recursive shortest spanning tree (RSST) algorithm. Then, all segment features are gathered together using a fuzzy multidimensional histogram to reduce the possibility of classifying similar segments to different classes. Extraction of several key frames is performed for each shot in a content-based rate-sampling framework. Two approaches are examined for key frame extraction. The first is based on examination of the temporal variation of the feature vector trajectory; the second is based on minimization of a cross-correlation criterion of the video frames. For efficient implementation of the latter approach, a logarithmic search (along with a stochastic version) and a genetic algorithm are proposed. Experimental results are presented which illustrate the performance of the proposed techniques, using synthetic and real life MPEG video sequences. (C) 1999 Academic Press.	en
heal.publisher	Acad Press Inc, Orlando, FL, United States	en
heal.journalName	Computer Vision and Image Understanding	en
dc.identifier.doi	10.1006/cviu.1999.0761	en
dc.identifier.isi	ISI:000082266200002	en
dc.identifier.volume	75	en
dc.identifier.issue	1	en
dc.identifier.spage	3	en
dc.identifier.epage	24	en