Fast forward to the early 2000s. Facial detection methodologies leaped forwards with Viola and Jones introducing a very fast method of detecting objects. The Viola-Jones method, while generic enough for the detection of any object, was primarily geared to detecting faces. The key genius to the Viola-Jones method is that it used many small classifiers to classify a region of an image, in a staged fashion. This is called the cascade classifier.
To make the explanation clearer, whenever classifier is used in the context of the Viola-Jones method, I mean the small classifiers in the cascade classifier. When referring to the cascade classifier, it will be explicitly mentioned as such.
A cascade classifier is made up of many small classifiers. ...
