PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Lp Norm Multiple Kernel Fisher Discriminant Analysis for Object and Image Categorisation
Fei Yan, Krystian Mikolajczyk, Mark Barnard, Hongping Cai and Josef Kittler
In: International Conference on Computer Vision and Pattern Recognition 2010, 13-18 Jun 2010, San Francisco, US.

Abstract

In this paper, we generalise multiple kernel Fisher discriminant analysis (MK-FDA) such that the kernel weights can be regularised with an Lp norm for any p>=1, in contrast to existing MK-FDA that uses either L1 or L2 norm. We present formulations for both binary and multiclass cases and solve the associated optimisation problems efficiently with semi-infinite programming. We show on three object and image categorisation benchmarks that by learning the intrinsic sparsity of a given set of base kernels using a validation set, the proposed Lp MK-FDA outperforms its fixed-norm counterparts, and is capable of producing state-of-the-art performance when applied to carefully designed base kernels. Moreover, we show that our Lp MK-FDA outperforms the Lp multiple kernel support vector machine (MK-SVM) which has been recently proposed. Based on this observation and our experience with single kernel FDA and SVM, we argue that the almost century-old FDA is still a strong competitor of the popular SVM.

EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
Learning/Statistics & Optimisation
Multimodal Integration
ID Code:6929
Deposited By:Fei Yan
Deposited On:22 April 2010