This paper proposes a method to search over a space of parameterized kernels using a gradient-descent based method.
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. The authors’ method effectively learns a non-linear representation of the data useful for classification and simultaneously performs dimensionality reduction. In addition, they suggest 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. (Published abstract provided)
810 Seventh Street NW, Washington, DC 20531, United States
Appears in 2007 IEEE Conference on Computer Vision and Pattern Recognition, Minneapolis, MN, USA, 2007, pp. 1-7