Description
Kernel estimation techniques, such as mean shift, suffer from one major drawback: the kernel bandwidth selection. The bandwidth can be fixed for all the data set or can vary at each points. The automatic bandwidth selection becomes a real challenge in case of multidimensional heterogeneous features. We have presented a solution to this problem. The selection is done iteratively for each type of features, by looking for the stability of local bandwidth estimates across a predefined range of bandwidths. The pseudo balloon mean shift filtering and partitioning are introduced. The validity of the method has been demonstrated in the context of color image segmentation based on 5-dimensional space.
References
A. Bugeau, P. 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, Troyes, September 2007.
(pdf)
A. Bugeau, P. Pérez. Detection and segmentation of moving objects in highly dynamic scenes.Technical report, INRIA, RR-6282, 2007.(pdf)