Face recognition is a very hot topic in Machine Learning. In this project, I will explore some existing methods on face recognition.
Image_proc is an simple example on how to process images.
The dataset I choose for face recognition is Yalefaces_A database. The database contains 165 GIF images of 15 subjects (subject01, subject02, etc.). There are 11 images per subject, one for each of the following facial expressions or configurations: center-light, w/glasses, happy, left- light, w/no glasses, normal, right-light, sad, sleepy, surprised, and wink.
First, I need to do is feature selection. I will try two different methods on facial features selection: PCA and ICA.
Then, I will use SVM and NN with different parameters to classify these faces.
PCA_SVM_ANN folder shows codes for PCA feature selection with SVM and ANN classification.
ICA_SVM_ANN folder shows codes of ICA feature selection with SVM and ANN classification.
Copy right by Charles Xu: charlesxu90@gmail.com