Learning Kernel Expansions for Image Classification

Abstract

Kernel machines (e.g. SVM, KLDA) have shown state-of-the-art performance in several visual classification tasks. The classification performance of kernel machines greatly depends on the choice of kernels and its parameters. In this paper, we propose a method to search over the space of parameterized kernels using a gradient-based method. Our method effectively learns a non-linear representation of the data useful for classification and simultaneously performs dimensionality reduction. In addition, we introduce a new matrix formulation that simplifies and unifies previous approaches. The effectiveness and robustness of the proposed algorithm is demonstrated in both synthetic and real examples of pedestrian and mouth detection in images.