Combined Classifiers for Invariant Face Recognition

TL;DR

This paper proposes a combined classifier system for face recognition that enhances stability and accuracy by integrating multiple classifiers, especially effective with unstable, low-performance classifiers, demonstrated on a dataset of 392 individuals.

Contribution

Introduces a novel system for combining unstable, low-performance classifiers to improve face recognition accuracy and stability.

Findings

High recognition rate achieved with fewer parameters

System demonstrates remarkable stability

Effective combination of classifiers enhances performance

Abstract

No single classifier can alone solve the complex problem of face recognition. Researchers found that combining some base classifiers usually enhances their recognition rate. The weaknesses of the base classifiers are reflected on the resulting combined system. In this work, a system for combining unstable, low performance classifiers is proposed. The system is applied to face images of 392 persons. The system shows remarkable stability and high recognition rate using a reduced number of parameters. The system illustrates the possibility of designing a combined system that benefits from the strengths of its base classifiers while avoiding many of their weaknesses.