Fast Bilateral Filtering of Vector-Valued Images

TL;DR

This paper introduces a fast algorithm for vector-valued bilateral filtering by approximating the Gaussian kernel with raised-cosines and employing Monte Carlo sampling, significantly speeding up the process while maintaining accuracy.

Contribution

The paper proposes a novel fast approximation method for vector-valued bilateral filtering using raised-cosine approximation and Monte Carlo sampling techniques.

Findings

Significant speedup over direct implementation

High accuracy in color image filtering

Effective approximation of Gaussian kernel

Abstract

In this paper, we consider a natural extension of the edge-preserving bilateral filter for vector-valued images. The direct computation of this non-linear filter is slow in practice. We demonstrate how a fast algorithm can be obtained by first approximating the Gaussian kernel of the bilateral filter using raised-cosines, and then using Monte Carlo sampling. We present simulation results on color images to demonstrate the accuracy of the algorithm and the speedup over the direct implementation.