People Monitoring Using Face Recognition With Observation Constraints (TA-P2)

Author(s) : Ji Tao (Nanyang Technological University, Singapore) Yap-Peng Tan (Nanyang Technological University, Singapore)

Abstract : We propose a novel probabilistic framework for recognizing people entering and leaving a closed-room in work place or living environment. In our method people are represented using facial feature, based on which optimal recognition is carried out by exploiting the temporal correlation among sequence of face observations. The optimality is achieved by maximizing a joint posterior probability. Experimental results are presented to show the efficacy of the proposed approach.

Menu