4.3. Location Estimation
In this section, we show that in the context of computer vision tasks, often only nonparametric approaches can provide a robust solution for the location estimation problem. We employ a class of nonparametric techniques in which the data points are regarded as samples from an unknown probability density. The location estimates are then defined as the modes of this density. Explicit computation of the density is avoided by using the mean shift procedure.
4.3.1. Why nonparametric methods?
The most general model of the location problem is that of multiple structures,
with no information being available about ...