Liveness detection in Face Recognition using Deep Learning
In the last few years face recognition has received a lot of attention. Recently face authentication has become very popular since it is easy to use and face recognition technologies have achieved good performance. Face biometric systems are widely used for recognizing the true identity of enrolled person on the basis of physiological appearances and behavioural patterns. However, this can be spoofed by nefarious users trying purposely to by-pass face recognition on system using various traits like photo or video of the person held in front of the camera may be accountable for performing face recognition. Using anti-face spoofing methods in our facial recognition framework, we tried to detect “real” or “fake” faces. By detecting possible spoofing attacks like placing 3D printed photo, eye & mouth photo imposter and video of a person in front of the camera, the proposed method improves liveness accuracy results.