March
2007
Parameterized Kernels for Support Vector Machine Classification
Authors:
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.
Notes:
associated project Component Analysis for Data Analysis
copied = false, 2000);
">
@conference{Frade-2007-9676,
author = {Fernando De La Torre Frade And Oriol Vinyals},
title = {Parameterized Kernels for Support Vector Machine Classification},
booktitle = {Proceedings of 2nd International Conference on Computer Vision Theory and Applications (VISAPP '07)},
year = {2007},
month = {March},
pages = {116 - 121},
keywords = {support vector machine, kernels, classification},
}
author = {Fernando De La Torre Frade And Oriol Vinyals},
title = {Parameterized Kernels for Support Vector Machine Classification},
booktitle = {Proceedings of 2nd International Conference on Computer Vision Theory and Applications (VISAPP '07)},
year = {2007},
month = {March},
pages = {116 - 121},
keywords = {support vector machine, kernels, classification},
}