Face detection method for color images

A method is proposed for the automatic detection of human faces from still images using color, shape and the gradient of luminance. The aim of the method is to detect faces having a great variety of sizes, positions and 3D orientations, including frontal and non-frontal views and situations when more than one face is present in the image. Candidate face regions are found by,combining color and shape criteria. For each candidate, parameters calculated from color, shape, and gradient of luminance features are used for evaluation. A measure of confidence that a region is face is defined using these parameters to allow tuning the face detector to application requirements.