Learning 2D Gabor Filters by Infinite Kernel Learning Regression
This work addresses image processing and computer vision tasks by providing a method to learn optimized Gabor filters for specific datasets, though it is incremental as it builds on existing kernel and sparse representation techniques.
The paper tackled the problem of image representation by proving 2D Gabor functions are translation-invariant positive-definite kernels and using infinite kernel learning regression to obtain a sparse representation via LASSO, leading to learned dataset-specific Gabor filters that improved face recognition accuracy.
Gabor functions have wide-spread applications in image processing and computer vision. In this paper, we prove that 2D Gabor functions are translation-invariant positive-definite kernels and propose a novel formulation for the problem of image representation with Gabor functions based on infinite kernel learning regression. Using this formulation, we obtain a support vector expansion of an image based on a mixture of Gabor functions. The problem with this representation is that all Gabor functions are present at all support vector pixels. Applying LASSO to this support vector expansion, we obtain a sparse representation in which each Gabor function is positioned at a very small set of pixels. As an application, we introduce a method for learning a dataset-specific set of Gabor filters that can be used subsequently for feature extraction. Our experiments show that use of the learned Gabor filters improves the recognition accuracy of a recently introduced face recognition algorithm.