Face Recognition from Still Images and Video

Face recognition with its wide range of commercial and law enforcement applications has been one of the most active areas of research in the field of computer vision and pattern recognition. Personal identification systems based on faces have the advantage that facial images can be obtained from a distance without requiring cooperation of the subject, as compared to other biometrics such as fingerprint, iris, etc. Face recognition is concerned with identifying or verifying one or more persons from still images or video sequences using a stored database of faces. Depending on the particular application, there can be different scenarios, ranging from controlled still images to uncontrolled videos. Since face recognition is essentially the problem of recognizing a 3D object from its 2D image or a video sequence, it has to deal with significant appearance changes due to illumination and pose variations. Current algorithms perform well in controlled scenarios, but their performance is far from satisfactory in uncontrolled scenarios. Most of the current research in this area is focused toward recognizing faces in uncontrolled scenarios. This chapter is broadly divided into two sections. The first section discusses the approaches proposed for recognizing faces from still images and the second section deals with face recognition from video sequences.