Face Recognition from Video: A CONDENSATION Approach

The aim of this work is to investigate how to exploit the temporal information in a video sequence for the task of face recognition. Inspired by Li and Chellappa's approach [11], we propose a probabilistic model parameterized by a tracking state vector and a recognizing identity variable, simultaneously characterizing the dynamics and identity of humans. We then invoke a CONDENSATION [8] approach to provide a numerical solution to the model. Once the joint posterior distribution of state vector and identity variable is estimated, we marginalize it over the state vector to yield a robust estimate of the posterior distribution of identity variable. Due to the propagation of identity and dynamics, a degeneracy in the posterior distribution of identity variable is achieved to give improved recognition. This evolving behavior is characterized using changes in entropy. The effectiveness of this approach is illustrated using experimental results on low-resolution video data