Semi-supervised learning with path-based similarity measure for face recognition

In this paper, we proposed a novel semi-supervised classification method with path-based similarity measure for face recognition. Based on the manifold assumption, our method can reflect genuine similarities between data points on manifolds without any other additional knowledge, which takes into account the existence of noise and outliers in the face dataset. Comparison experiments between the proposed method and the other two methods: PCA and LDA, are performed. The results show that the proposed method achieves the best face recognition.