Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.contributor.authorKaravasilis, V.en
dc.contributor.authorNikou, C.en
dc.contributor.authorLikas, A.en
dc.rightsDefault Licence-
dc.subjectvisual trackingen
dc.subjectgaussian mixture model (gmm)en
dc.subjectexpectation-maximization (em) algorithmen
dc.subjectdifferential earth mover's distance (differential emd)en
dc.subjectkalman filteren
dc.subjectobject trackingen
dc.titleVisual tracking using the Earth Mover's Distance between Gaussian mixtures and Kalman filteringen
heal.type.enJournal articleen
heal.type.elΆρθρο Περιοδικούel
heal.identifier.primaryDOI 10.1016/j.imavis.2010.12.002-
heal.recordProviderΠανεπιστήμιο Ιωαννίνων. Σχολή Θετικών Επιστημών. Τμήμα Μηχανικών Ηλεκτρονικών Υπολογιστών και Πληροφορικήςel
heal.abstractIn this paper, we demonstrate how the differential Earth Mover's Distance (EMD) may be used for visual tracking in synergy with Gaussian mixtures models (GMM). According to our model, motion between adjacent frames results in variations of the mixing proportions of the Gaussian components representing the object to be tracked. These variations are computed in closed form by minimizing the differential EMD between Gaussian mixtures, yielding a very fast algorithm with high accuracy, without recurring to the EM algorithm in each frame. Moreover, we also propose a framework to handle occlusions, where the prediction for the object's location is forwarded to an adaptive Kalman filter whose parameters are estimated on line by the motion model already observed. Experimental results show significant improvement in tracking performance in the presence of occlusion. (C) 2010 Elsevier B.V. All rights reserved.en
heal.journalNameImage and Vision Computingen
heal.journalTypepeer reviewed-
Appears in Collections:Άρθρα σε επιστημονικά περιοδικά ( Ανοικτά)

Files in This Item:
File Description SizeFormat 
Karavasilis-2011-Visual tracking usin.pdf2.41 MBAdobe PDFView/Open    Request a copy

This item is licensed under a Creative Commons License Creative Commons