Eigenfaces for Recognition

TLDR

The system’s approach is motivated by physiology, information theory, and the need for near‑real‑time accuracy, using eigenfaces—principal‑component features of face images that need not correspond to specific facial parts. It projects face images onto a low‑dimensional eigenface feature space, representing each face by a weighted sum of eigenfaces and recognizing individuals by comparing these weights, treating recognition as a 2‑D problem without 3‑D reconstruction. The resulting system can locate, track, and recognize faces in near‑real‑time, learns new faces unsupervised, and is easily implemented with a neural network.

Abstract

We have developed a near-real-time computer system that can locate and track a subject's head, and then recognize the person by comparing characteristics of the face to those of known individuals. The computational approach taken in this system is motivated by both physiology and information theory, as well as by the practical requirements of near-real-time performance and accuracy. Our approach treats the face recognition problem as an intrinsically two-dimensional (2-D) recognition problem rather than requiring recovery of three-dimensional geometry, taking advantage of the fact that faces are normally upright and thus may be described by a small set of 2-D characteristic views. The system functions by projecting face images onto a feature space that spans the significant variations among known face images. The significant features are known as "eigenfaces," because they are the eigenvectors (principal components) of the set of faces; they do not necessarily correspond to features such as eyes, ears, and noses. The projection operation characterizes an individual face by a weighted sum of the eigenface features, and so to recognize a particular face it is necessary only to compare these weights to those of known individuals. Some particular advantages of our approach are that it provides for the ability to learn and later recognize new faces in an unsupervised manner, and that it is easy to implement using a neural network architecture.