Briefly describe the process of face recognition using Principal Components Analysis (Eigenfaces). In your answer you should address the following issues a) Why is it important to align all the face images and remove the b) Why is it impossible to use Principal Components Analysis directly on the c) What is the 'trick' used in Eigenface-based face-recognition to reduce the d) What are the problems with holistic techniques such as PCA that use the background? covariance matrix of the face images? complexity of the eigenvector decomposition? whole face as input compared to component techniques which process the face as multiple regions