Facial Recognition Project 1

The project was implemented using Matlab programming. The provided face data were raster-scanned face images (46x56 pixels) in columns.

Section 1 - Computationally Efficient Eigenfaces

In this section the provided face data are partitioned into your training and testing data. Then PCA is applied the training data and the a discussion is followed. Face image reconstruction is implemented varying the number of PCA bases learnt firstly and secondly using PCA-based face recognition by the NN classification.

Section 2 - Incremental PCA

Incremental PCA is performed and the a discussion is followed answering how accurate the incremental method is, what are the important parameters in the method.

Section 3 - PCA-LDA Ensemble for Face Recognition

Firslty, the PCA-LDA based face recognition with the NN classifier is performed and then the PCA-LDA Ensemble method followed by a discussion.