Learning OpenCV

Face Detection or Haar Classifier

We now turn to the final tree-based technique in OpenCV: the Haar classifier, which builds a boosted rejection cascade. It has a different format from the rest of the ML library in OpenCV because it was developed earlier as a full-fledged face-recognition application. Thus, we cover it in detail and show how it can be trained to recognize faces and other rigid objects.

Computer vision is a broad and fast-changing field, so the parts of OpenCV that implement a specific technique—rather than a component algorithmic piece—are more at risk of becoming out of date. The face detector that comes with OpenCV is in this "risk" category. However, face detection is such a common need that it is worth having a baseline technique that works fairly well; also, the technique is built on the well-known and often used field of statistical boosting and thus is of more general use as well. In fact, several companies have engineered the "face" detector in OpenCV to detect "mostly rigid" objects (faces, cars, bikes, human body) by training new detectors on many thousands of selected training images for each view of the object. This technique has been used to create state-of-the-art detectors, although with a different detector trained for each view or pose of the object. Thus, the Haar classifier is a valuable tool to keep in mind for such recognition tasks.

OpenCV implements a version of the face-detection technique first developed by Paul Viola and Michael Jones—commonly ...