EXAMPLE-BASED LEARNING FOR VIEW-BASED HUMAN FACE DETECTION

We present an example-based learning approach for locating vertical fr ontal views of human faces in complex scenes. The technique models the distribution of human face patterns by means of a few view-based ''fa ce'' and ''nonface'' model clusters. At each image location, a differe nce feature vector is computed between the local image pattern and the distribution-based model. A trained classifier determines, based on t he difference feature vector measurements, whether or not a human face exists at the current image location. We show empirically that the di stance metric we adopt for computing difference feature vectors, and t he ''nonface'' clusters we include in our distribution-based model, ar e both critical for the success of our system.