Face recognition using improved pairwise coupling support vector machines

A novel structure is proposed to tackle multiclass classification problem. For a K-class classification task, an array of K optimal pairwise coupling classifier (O-PWC) is constructed, each of which is the most reliable and optimal for the corresponding class in the sense of cross entropy or square error. The final decision will be got through combining the results of these K OPWCs. The accuracy rate is improved while the computational cost will not increase too much. This algorithm is applied to face recognition on Cambridge ORL face database, the experimental results reveal our method is effective and efficient.