As we have discussed briefly in Chapter 7, Extracting Image Features and Descriptors, (in the context of Haar-like feature extraction), the Viola-Jones' object detection technique can be used for face detection in images. It is a classical machine learning approach, where a cascade function is trained using a training set of positive and negative images, by using the hand-crafted Haar-like features extracted from the images.
The Viola-Jones algorithm typically uses a base patch size (for example, 24 x 24 pixels) that it slides across and down an image and computes a huge number of Haar-like features (160,000 possible features for a 24 x 24 patch, although ...