Catalogue of Artificial Intelligence Techniques


Jump to: Top | Entry | References | Comments

View Maths as: Images | MathML


Aliases: Eigenimage, Face Recognition

Keywords: face, face recognition, image recognition

Categories: Pattern Recognition and Image Processing

Author(s): Andrey Finayev

Problem Statement

Train a system with various types of human faces so that it can recognize a face from a new randomly chosen picture.

Definition of technique

Eigenfaces are a set of eigenvectors used in the computer vision problem of human face recognition. These eigenvectors can be derived from the covariance matrix of the probability distribution of the high-dimensional vector space of possible faces of human beings. Eigenface technique is also reffered to as eigenimage, as the technique has been used for handwriting, lip reading, voice recognition, and medical imaging.


The eigenface face recognition system can be divided into two main segments: creation of the eigenface basis and recognition, or detection, of a new face. The system follows the following general flow:

Eigenfaces are a set of "standardized face ingredients", derived from statistical analysis of many pictures of faces. One sphere, where eigenface technique shows itself very useful is face recognition security software. Any human face can be considered to be a combination of standard faces. It's possible to recreate a human face usign 10% of information from face 1, 24% from face 2 and so on. This means that if you want to record someone's face for use by face recognition software you can use less space than would be taken up by a digitised photograph.

Some examples of eigenfaces from the database:

To generate a set of eigenfaces, a large set of digitized images of human faces, taken under the same lighting conditions, are normalized to line up the eyes and mouths. They are then all resampled at the same pixel resolution (say m×n), and then treated as mn-dimensional vectors whose components are the values of their pixels. The eigenvectors of the covariance matrix of the statistical distribution of face image vectors are then extracted. It should be noted that these are the same as the eigenvectors from principal components analysis, the statistical method from which eigenimaging is derived.

Since the eigenvectors belong to the same vector space as face images, they can be viewed as if they were m×n pixel face images: hence the name eigenfaces.

When properly weighted, eigenfaces can be summed together to create an approximate gray-scale rendering of a human face. Remarkably few eigenvector terms are needed to give a fair likeness of most people's faces, so eigenfaces provide a means of applying data compression to faces for identification purposes.



Add Comment

No comments.