Blinking-based live face detection using conditional random fields

Abstract

This paper presents a blinking-based liveness detection method for human face using Conditional Random Fields (CRFs). Our method only needs a web camera for capturing video clips. Blinking clue is a passive action and does not need the user to to any hint, such as speaking, face moving. We model blinking activity by CRFs, which accommodates long-range contextual dependencies among the observation sequence. The experimental results demonstrate that the proposed method is promising, and outperforms the cascaded Adaboost method and HMM method. © Springer-Verlag Berlin Heidelberg 2007.