Fast High-Dimensional Kernel Filtering

Nair, Pravin and Chaudhury, Kunal N (2019) Fast High-Dimensional Kernel Filtering. In: IEEE SIGNAL PROCESSING LETTERS, 26 (2). pp. 377-381.

PDF Iee_Sig_Pro_Let_26-2_377_2019.pdf - Published Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

The bilateral and nonlocal means filters are instances of' kernel-based filters that are popularly used in image processing. It was recently shown that fast and accurate bilateral filtering of grayscale images can he performed using a low-rank approximation of the kernel matrix. More specifically, based on the eigendecomposition of the kernel matrix, the overall filtering was approximated using spatial convolutions, for which efficient algorithms are available. Unfortunately, this technique cannot be scaled to high-dimensional data such as color and hyperspectral images. This is simply because one needs to compute/store a large matrix and perform its eigendecomposition in this case. We show how this problem can be solved using the Nystrom method, which is generally used for approximating the eigendecomposition of large matrices. The resulting algorithm can also be used for nonlocal means filtering. We demonstrate the effectiveness of our proposal for bilateral and nonlocal means filtering of color and hyperspectral images. In particular, our method is shown to be competitive with state-of-the-art fast algorithms, and moreover, it comes with a theoretical guarantee on the approximation error.

Item Type:	Journal Article
Publication:	IEEE SIGNAL PROCESSING LETTERS
Publisher:	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Additional Information:	Copyright of this article belongs to IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Keywords:	Kernel filter; Nystrom method; approximation; fast algorithm; error bound
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	20 Feb 2019 04:59
Last Modified:	01 Mar 2019 07:05
URI:	http://eprints.iisc.ac.in/id/eprint/61760

Actions (login required)

View Item