Python implementation of the eigenfaces technique

I provide a Python implementation of the eigenfaces technique for representing real facial images using an assortment of base images (computed by performing an eigenanalysis of the face data matrix).

You can use the code with a variety of aligned face image data sets, such as the FERET data set of faces.

First, throw all your aligned face images in "fas". Then run "python eigenfaces.py". Wait until the script terminates, and then look into your current directory.

I have obtained and used the FERET data set of face images while writing this script. Due to licensing issues, I can not include any samples of this data set in this code repository.

You are advised to obtain FERET on your own; see [1] for more info.

Please direct all your input to sokar@aiia.csd.auth.gr

References

[1] "FERET FAQ", http://www.nist.gov/itl/iad/ig/colorferet.cfm