Class-Specific Nonlinear Projections Using Class-Specific Kernel Spaces

In this paper, we propose a new approach for nonlinear Class-specific Discriminant Analysis that exploits a class-specific kernel space definition. We show that the proposed method can considerably reduce the time and space complexities of the standard Class-specific Kernel Discriminant Analysis. Our analysis is verified by experiments illustrating the efficiency of the proposed class-specific kernel-based learning approach.