Skip to Main content Skip to Navigation
Conference papers

Sélection de la taille du noyau pour l'estimation à noyau dans des espaces multidimensionnels hétérogènes.

Aurélie Bugeau 1 Patrick Pérez 1
1 VISTA - Vision spatio-temporelle et active
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Kernel estimation techniques, such as mean shift, suffer from one major drawback: the kernel bandwidth selection. This selection becomes a real challenge in case of multidimensional heterogeneous features. This paper presents a solution to this problem. The selection is done iteratively for each type of features, by looking for the stability of local bandwidth estimates within a predefined range of bandwidths. A new estimator that permits the iterative computation is introduced. The validity of the method is demonstrated in the context of color image segmentation and motion segmentation.
Document type :
Conference papers
Complete list of metadata

Cited literature [10 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00551598
Contributor : Aurélie Bugeau <>
Submitted on : Tuesday, January 4, 2011 - 10:39:59 AM
Last modification on : Thursday, March 25, 2021 - 3:31:37 AM
Long-term archiving on: : Tuesday, April 5, 2011 - 2:47:20 AM

File

Gretsi_final.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00551598, version 1

Citation

Aurélie Bugeau, Patrick Pérez. Sélection de la taille du noyau pour l'estimation à noyau dans des espaces multidimensionnels hétérogènes.. 21ème colloque GRETSI sur le traitement du signal et des images, 2007, France. ⟨hal-00551598⟩

Share

Metrics

Record views

280

Files downloads

166