Face Detection Using a Boosted Cascade of Features Using OpenCV

Abstract

Face detection in an image is a problem that has gained lots of importance in the last decade. Detecting faces is quite simple for human beings because it comes naturally but it is not so easy to teach a computer to detect faces. We divide the detection problem into three steps. The first step is Integral Image [1] which allows the features used by detector to be computed very quickly. The second step is a learning algorithm, based on AdaBoost, which selects a small number of critical visual features from a larger set and yields extremely efficient classifiers [2]. The third step is combining more complex classifiers in a cascade [3] which allows background regions of the image to be quickly discarded while spending more computation on promising face like regions. We conduct extensive experiments on publicly available databases to verify the efficacy of the proposed algorithm. In real time the detector runs at 15 frames per second without resorting to image difference or skin color detection.