Face recognition in a transformed domain

TL;DR

This paper explores using the discrete cosine transform (DCT) for face recognition, demonstrating improved accuracy and efficiency over traditional eigenfaces methods through various classifier combinations.

Contribution

It introduces DCT as an alternative to eigenfaces for face recognition, showing enhanced performance and reduced computational load.

Findings

DCT-based recognition outperforms eigenfaces in accuracy

Reduced computational burden with DCT

Effective with multiple classifiers

Abstract

This paper proposes the use of a discrete cosine transform (DCT) instead of the eigenfaces method (Karhunen-Loeve Transform) for biometric identification based on frontal face images. Experimental results show better recognition accuracies and reduced computational burden. This paper includes results with different classifiers and a combination of them.