An Independent Evaluation of Subspace Face Recognition Algorithms

TL;DR

This study compares linear and kernel-based face recognition algorithms across various challenging conditions, evaluates their performance with statistical tools, and explores hybrid fusion strategies to improve recognition accuracy.

Contribution

It provides an independent, comprehensive evaluation of PCA, LDA, and ICA methods under diverse conditions and introduces a hybrid approach combining the best algorithms.

Findings

Kernel methods outperform linear counterparts in most scenarios.

Fusion strategies enhance recognition robustness and accuracy.

Evaluation framework facilitates fair comparison of face recognition algorithms.

Abstract

This paper explores a comparative study of both the linear and kernel implementations of three of the most popular Appearance-based Face Recognition projection classes, these being the methodologies of Principal Component Analysis, Linear Discriminant Analysis and Independent Component Analysis. The experimental procedure provides a platform of equal working conditions and examines the ten algorithms in the categories of expression, illumination, occlusion and temporal delay. The results are then evaluated based on a sequential combination of assessment tools that facilitate both intuitive and statistical decisiveness among the intra and interclass comparisons. The best categorical algorithms are then incorporated into a hybrid methodology, where the advantageous effects of fusion strategies are considered.