Although there are many works that have been done in face tracking, most of them are not suitable for realtime requirement in smart phones which have computation limitations. In the other hand, simple algorithms capable of applying for mobile devices still have some problems in tracking. With the combination approach, we propose a face tracking system that is fast, accuracy, and easy to implement. Our tracking system utilizes the particle filter and optical flow frameworks. By incorporating the strong points of each algorithm, the tracking process not only becomes more robust but also satisfies the realtime constraint. The result of our method in the smart phone shows an attractive performance and feasibility for future applications.