Face recognition is a method that uses digital images or video frames of human faces to identify or verify the identity of a person. It has been a critical human computer interaction technique and has been widely applied in plenty of areas, such as video surveillance, identity verification, access control, public security system, and even daily life.
Face recognition is a persistent topic. Traditional methods are based on shallow learning. This kind of methods only use basic features of images, which have been facing many challenges, such as lighting, shadow, complexity of background, facial expression, and pose variation. However, Deep learning has changed this situation. Deep learning-based methods can extract more complicated features and it is more accurate and stable.