Face verification via error correcting output codes

doi:10.1016/j.imavis.2003.09.013

2003-12-01

PRE

Reza Ghaderi

Terry Windeatt

Jiřı́ Matas

DOI：10.1016/j.imavis.2003.09.013

原文链接

原文求助

摘要

En 中文

We propose a novel approach to face verification based on the Error Correcting Output Coding (ECOC) classifier design concept. In the training phase, the client set is repeatedly divided into two ECOC specified sub-sets (super-classes) to train a set of binary classifiers. The output of the classifiers defines the ECOC feature space, in which it is easier to separate transformed patterns representing clients and impostors. As a matching score in this space, we propose the average first order Minkowski distance between the probe and gallery images. The proposed method exhibits superior verification performance on the well known XM2VTS data set as compared with previously reported results. (C) 2003 Elsevier B.V. All rights reserved.

Keyword:

error correcting output coding

sub-sets

Minkowski