Face recognition technologies for evidential evaluation of video traces

Human recognition from video traces is an important task in forensic investigations and evidence evaluations. Compared with other biometric traits, face is one of the most popularly used modalities for human recognition due to the fact that its collection is non-intrusive and requires less cooperation from the subjects. Moreover, face images taken at a long distance can still provide reasonable resolution, while most biometric modalities, such as iris and fingerprint, do not have this merit. In this chapter, we discuss automatic face recognition technologies for evidential evaluations of video traces. We first introduce the general concepts in both forensic and automatic face recognition, then analyse the difficulties in face recognition fromvideos.We summarise and categorise the approaches for handling different uncontrollable factors in difficult recognition conditions. Finally we discuss some challenges and trends in face recognition research in both forensics and biometrics. Given its merits tested in many deployed systems and great potential in other emerging applications, considerable research and development efforts are expected to be devoted in face recognition in the near future.