Real time face detection and recognition
Evangelos Sarıyanidi, Volkan Dağlı, Salih Cihan Tek, Birkan Tunç, Muhittin Gökmen · 2012
In this demo session, a real-time automatic face detection and recognition system will be demonstrated. The system, which is implemented as a desktop application with a user interface, detects the faces in the images that are grabbed from a web camera using a cascaded classifier consisting of Modified Census Transform features. Then, using the same method, it locates the eyes and the mouth on each face and uses this information to align the faces. Finally, it recognizes these aligned faces using a novel method called local Zernike moments. In order to improve the detection performance, the system also includes face tracking. The result of the recognition process can be observed by the names that are written by the system near the faces. The system makes it possible to add a previously unseen person to the database easily via its user interface.