Home | Intro | Proposal | Algorithmics | Results | Code | Images | References

Eigenfaces Group - Algorithmics

Constructing Eigenfaces

This procedure is a form of principle component analysis. First, the conceptually simple version:

Unfortunately, this procedure relies on computing eigenvectors of an extremely large matrix. Our images are 250x300, so the matrix would be 75000 by 75000 (5.6 billion entries!). On the bright side, there's another way (the Karhunen-Loève expansion):

Transforming an image to face space

This procedure is exactly what you would expect for the usual Hilbert space change of basis.

The inverse face space transform

Again, just what you would expect.

"Learning" a face

Recognizing a known face

Evaluating "face-ness" of an image

This procedure could also serve for searching for faces in a larger image. At least, that's what the MIT people us it for.

Tim Danner <tdanner@rice.edu>, Indraneel Datta <kashent@rice.edu>
Last modified: Fri Dec 17 20:43:02 CST 1999