A SUPPORT VECTOR MACHINE BASED DYNAMIC CLASSIFIER FOR FACE RECOGNITION

摘要

Most of the researches on support vector machine (SVM) based face recognition presume that the classifier, once trained, is static and thus unscalable, due to the fact that SVM is a supervised learning method. This paper introduces a novel SVM-based face recognition method, which circumvents this difficulty, by allowing "new" faces of existing or new persons to be added into the face database dynamically. In other words, the proposed method is capable of learning and recognizing faces that are not already in the face database. Our experimental results indicate that the accuracy rate of the proposed method ranges from 73.81% up to 100% and outperforms all the methods we evaluated. Moreover, this paper uses several different tests to analyze the performance of the proposed algorithm.