We use state-of-the-art deep learning algorithms to do face recognition. Each new person needs to have his or her face encoded and added to the database. The face recognition software will take images of the new person’s face, apply convolutional and non-linear filters repeatedly on them. It could go anywhere from 10 layers to 100 layers, depending on the needs. In so doing, it will extract higher order information, and the face is encoded into a small template. During face recognition, the person’s face is processed and checked against existing template for match.