jaimeivancervantes/facerecognition

Face recognition written in MATLAB. This code implements Fisher's Discriminant Analysis and SVMs using LIBSVM.

facerecognition's Issues

The question about the PCA part in the "main.m"

I read your code carefully and I have a question about the PCA part in the "main.m".
As I know, to play the PCA on "A", we need to get the eigenvectors of "A^T*A". And these sorted eigenvectors form the loading matrix. I don't know why you get the eigenvectors as follows:
% Obtaining the eigenvectors U = A * pevec;
Because I think the pevec is already the eigenvectors or the loading matrix.

Running the code with other database

I am trying to run the SVM part with images 480x640 pixels and the acuraccy is zero.When i resize the images to be 48x40 pixels as in your code the acuraccy is slightly better.Is there any way that i can achieve 100% accuracy?

Using with MULTI-PIE dataset

Is there any way to use this code with Multi PIE dataset?