Confidence Measure for Automatic Face Recognition

Confidence Measure for Automatic Face Recognition

This paper deals with the use of confidence measure for Automatic Face Recognition (AFR). AFR is realizedby the adapted Kepenecki face recognition approach based on the Gabor wavelet transform. This work ismotivated by the fact that obtained recognition rate on the real-world corpus is only about 50% which is notsufficient for our application, a system for automatic labelling of the photographs in a large database. Themain goal of this work is thus the proposition of the post-processing of the classification result in order toremove the “incorrectly” classified face images. We show that the use of confidence measure to filter outincorrectly recognized faces is beneficial. Two confidence measures are proposed and evaluated on the Czech News Agency (ˇCTK) corpus. Experimental results confirm the benefit of the use of confidence measure forthe automatic face recognition task.

Keywords: Automatic Face Recognition; Confidence Measure; Gabor Wavelets; Czech News Agency

Year: 2011

Download: download Full text [60 kB]

Authors of this publication:

Pavel Král

Phone: +420 377 632 454
E-mail: pkral@kiv,

Pavel is a lecturer/researcher at the Department of Computer Science and Engineering at the University of West Bohemia in Pilsen (Czech Republic). His research is focused on automatic speech processing, dialog act recognition, syntactic parsing, punctuation annotation and document classification.