Face spoofing detection
Summarization
Face recognition is one of the most commonly used techniques in applications of biometrics, e.g. access control, law enforcement, multimedia communication, human-computer interaction. Like other biometric modalities, however, a face recognition system can be attacked easily and at very low cost by two common attacks, namely print attack and replay attack. In print attacks, face spoofing is carried out by presenting a printed photo to a camera. In replay attacks, on the other hand, the attackers replay a previously recorded face image or video of a targeted user in order to spoof the biometric system. As attackers only need to acquire a printed photo or a video of the authorized user's face, with current technologies these types of attack can be carried out easily in both remote and logical access control systems protected by a face recognition system.
We propose an approach for face spoofing detection using the high-order Local Derivative Pattern from Three Orthogonal Planes (LDP-TOP). The proposed method is not only simple to derive and implement, but also highly efficient, since it takes into account both spatial and temporal information in different directions of subtle face movements.
References
I am Quoc Tin Phan, the 3rd-year PhD student of the University of Trento, Italy. My research interest is multimedia forensics and applied machine learning.
quoctin.phan@unitn.it