Logo des Repositoriums
 

Superficial Gaussian mixture reduction

dc.contributor.authorHuber, Marco F.
dc.contributor.authorKrauthausen, Peter
dc.contributor.authorHanebeck, Uwe D.
dc.contributor.editorHeiß, Hans-Ulrich
dc.contributor.editorPepper, Peter
dc.contributor.editorSchlingloff, Holger
dc.contributor.editorSchneider, Jörg
dc.date.accessioned2018-11-27T10:00:08Z
dc.date.available2018-11-27T10:00:08Z
dc.date.issued2011
dc.description.abstractMany information fusion tasks involve the processing of Gaussian mixtures with simple underlying shape, but many components. This paper addresses the problem of reducing the number of components, allowing for faster density processing. The proposed approach is based on identifying components irrelevant for the overall density's shape by means of the curvature of the density's surface. The key idea is to minimize an upper bound of the curvature while maintaining a low global reduction error by optimizing the weights of the original Gaussian mixture only. The mixture is reduced by assigning zero weights to reducible components. The main advantages are an alleviation of the model selection problem, as the number of components is chosen by the algorithm automatically, the derivation of simple curvature-based penalty terms, and an easy, efficient implementation. A series of experiments shows the approach to provide a good trade-off between quality and sparsity.en
dc.identifier.isbn978-88579-286-4
dc.identifier.pissn1617-5468
dc.identifier.urihttps://dl.gi.de/handle/20.500.12116/18805
dc.language.isoen
dc.publisherGesellschaft für Informatik e.V.
dc.relation.ispartofINFORMATIK 2011 – Informatik schafft Communities
dc.relation.ispartofseriesLecture Notes in Informatics (LNI) - Proceedings, Volume P-192
dc.titleSuperficial Gaussian mixture reductionen
dc.typeText/Conference Paper
gi.citation.endPage491
gi.citation.publisherPlaceBonn
gi.citation.startPage491
gi.conference.date4.-7. Oktober 2011
gi.conference.locationBerlin
gi.conference.sessiontitleRegular Research Papers

Dateien

Originalbündel
1 - 1 von 1
Lade...
Vorschaubild
Name:
491.pdf
Größe:
20.98 KB
Format:
Adobe Portable Document Format