Interocular Distance based Facial Recognition

Facial Recognition has become a key feature when it comes to biometrics. In order to capture a person's face, there are certain vital attributes of the face that we need to understand such as length and width of nose, eyes, chin, eyebrows etc. There are many existing methods (algorithms) for performing facial detection but it is difficult to assess the performance of these methods because of its complexity. In this paper, facial recognition using the property of golden ratio of human face is discussed. The proposed technique is more efficient in terms of run time length and detection process is easy as it only stores the eye data. Hence measuring only the distance between the eyes would help us do facial recognition. A realistic comparison between Haar feature-based cascade classifier and contour mapping based algorithm is presented. Georgia tech university's dataset has been used for the purpose of comparison of different algorithms.