Classifier fusion of improved Laplacianfaces

This paper presents a new face recognition method for human-robot interaction. Laplacianface is recently proposed, which can find an embedding subspace that preserves local structure information, and can deal with new test data easily. However, it has limitations in recognition tasks for not taking class-specific information into account. A novel face recognition approach is proposed based on the combination of the Gabor wavelet, Kernel Discriminant Laplacianfaces (KDL) and classifier fusion. The experiments show the effectiveness and encouraging performance compared with some other existing methods.