Biometric liveness detection based on cross modal fusion

In this paper we propose liveness checking technique for multimodal biometric authentication systems based on audio-visual cross-modal fusion. Liveness checking ensures that biometric cues are acquired from a live person who is actually present at the time of capture for authenticating the identity. The liveness check based on mutual dependency models is performed by fusion of acoustic and visual speech features which measure the degree of synchrony between the lips and the voice extracted from speaking face video sequences. Performance evaluation in terms of DET (detector error tradeoff) curves and EERs(equal error rates) on publicly available audiovisual speech databases show a significant improvement in performance of proposed fusion of face-voice features based on mutual dependency models.