PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Using the Fisher-Rao Metric to Compute Facial Similarity
Simone Ceolin and Edwin Hancock
In: International Conference on Image Analysis and Recognition, June 21-23, 2010, Povoa de Varzim, Portugal.

Abstract

In this paper we show how the Fisher-Rao metric can be used to compute the similarity of fields of surface normals, under the assumption of a von-Mises Fisher (vMF) distribution. We use the similarity measure to analyse differences in facial shape due to gender and expression. Finally, we show the results achieved using BU-3DFEDB and Max Planck datasets.

EPrint Type:Conference or Workshop Item (Oral)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
Learning/Statistics & Optimisation
ID Code:7385
Deposited By:Edwin Hancock
Deposited On:17 March 2011

[Edit]