We are now ready to build a face recognizer. We need a face dataset for training, so we've provided you with a folder called faces_dataset that contains a small number of images that are sufficient for training. This dataset is a subset of the dataset that is available at http://www.vision.caltech.edu/Image_Datasets/faces/faces.tar. This dataset contains a good number of images that we can use to train a face recognition system.
We will use a local binary patterns histogram to build our face recognition system. In our dataset, you will see different people. Our job is to build a system that can learn to separate these people from one another. When we see an unknown image, our system will assign it to one of the existing classes.