Improved Eigenfeature Regularization for Face Identification

TL;DR

This paper introduces a novel eigenspectrum modeling approach that divides each face class into subclasses to better capture intra-personal variations, leading to improved face recognition performance.

Contribution

It proposes a new method of eigenspectrum modeling combined with subclass division to enhance discriminative feature extraction for face identification.

Findings

Outperforms existing methods on AR, FERET, and YouTube Face databases.

Effectively captures intra-personal variances for better recognition.

Provides a comprehensive analysis of eigenspectrums in face images.

Abstract

In this work, we propose to divide each class (a person) into subclasses using spatial partition trees which helps in better capturing the intra-personal variances arising from the appearances of the same individual. We perform a comprehensive analysis on within-class and within-subclass eigenspectrums of face images and propose a novel method of eigenspectrum modeling which extracts discriminative features of faces from both within-subclass and total or between-subclass scatter matrices. Effective low-dimensional face discriminative features are extracted for face recognition (FR) after performing discriminant evaluation in the entire eigenspace. Experimental results on popular face databases (AR, FERET) and the challenging unconstrained YouTube Face database show the superiority of our proposed approach on all three databases.