Evolving ensemble classifiers for incremental face recognition

Face recognition is an important research issue in many modern applications. Traditional face recognition technologies need collect the set of enough face images to learn an accurate face classifier for a specific person. However, collecting face images at a time is not easy for most people in real applications. In this paper, we propose an incremental learning method based on data selection and random decision trees to learn ensemble classifiers for on-line face recognition. The experimental results show that the proposed incremental learning method is efficient and highly adaptive. The learned classifiers also have accurate and stable recognition rate under a limited number of training face datasets.