KIT | KIT-Bibliothek | Impressum | Datenschutz

Paper G. Superficial Gaussian Mixture Reduction. Edited version of the paper: M. F. Huber, P. Krauthausen, and U. D. Hanebeck. Superficial Gaussian Mixture Reduction. In INFORMATIK 2011 - the 41th Annual Conference of the Gesellschaft für Informatik e.V. (GI), 6thWorkshop Sensor Data Fusion: Trends, Solutions, Applications (SDF), Berlin, Germany, October 2011

Huber, Marco F.; Krauthausen, Peter; Hanebeck, Uwe D.

Many 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 algorithmautomatically, 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.

Volltext §
DOI: 10.5445/IR/1000046060
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Anthropomatik und Robotik (IAR)
Publikationstyp Buchaufsatz
Publikationsjahr 2015
Sprache Englisch
Identifikator urn:nbn:de:swb:90-460687
KITopen-ID: 1000046068
Erschienen in Nonlinear Gaussian Filtering : Theory, Algorithms, and Applications. Ed.: M. Huber
Verlag Karlsruher Institut für Technologie (KIT)
Seiten 360-377
Relationen in KITopen
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page