Face detection using the Haar cascade

Paul Viola and Micheal Jones proposed the Haar feature-based cascade classifier in their paper, Rapid Object Detection using a Boosted Cascade of Simple Features, in 2001. The Haar feature-based cascade classifier is trained using facial images as well as non-facial images. The Haar cascade classifier can not only detect a frontal face but can also detect the eyes, mouth, and nose of a person. The Haar feature-based classifier is also referred to as the Viola-Jones algorithm.

Basic working of the Viola-Jones algorithm

So, put simply, the Viola-Jones algorithm used Haar features to detect a face. Haar generally consists of two main features: edge features and line features. We will first...