Kernel Density Estimation on Spaces of Gaussian Distributions and Symmetric Positive Definite Matrices - Archive ouverte HAL Access content directly
Journal Articles SIAM Journal on Imaging Sciences Year : 2017

Kernel Density Estimation on Spaces of Gaussian Distributions and Symmetric Positive Definite Matrices

Abstract

This paper analyzes the kernel density estimation on spaces of Gaussian distributions endowed with different metrics. Expressions of kernels are provided for the 2-Wasserstein metric on the space of multivariate Gaussians. For the Fisher metric the kernels are provided only for univariate Gaussians and multivariate centered Gaussians. The density estimation is successfully applied to a classification problem of electro-encephalographic signals.
Fichier principal
Vignette du fichier
KernelDensityEstimationOnGaussianDistributions.pdf (2.53 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01535731 , version 1 (09-06-2017)

Identifiers

Cite

Emmanuel Chevallier, Emmanuel Kalunga, Jesus Angulo. Kernel Density Estimation on Spaces of Gaussian Distributions and Symmetric Positive Definite Matrices. SIAM Journal on Imaging Sciences, 2017, 10 (1), pp.191 - 215. ⟨10.1137/15M1053566⟩. ⟨hal-01535731⟩
4750 View
199 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More