The method of template matching consists of directly comparing the appearance of a given image with a reference image by means of a suitable metric. To extract features of interest (eyes, mouth) from the data, a parameterized template can be used [153]. A prior knowledge of feature shape, such as the eye, is used in an energy minimizing scheme [19][154][155][48]. These templates are specified by a set of parameters which enable a priori knowledge about the expected shape of a feature to guide the detection process. The deformable templates interact with the image in a dynamic manner. An energy function is defined. It contains terms attracting the template to salient features (intensity, peaks and valleys, edges). The parameters are obtained by a minimization of the energy function. This method is relatively insensitive to variations in scale, tilt and rotation of the head, and to lighting conditions.