CVMay 7, 2016

Fast Bilateral Filtering of Vector-Valued Images

arXiv:1605.02164v118 citations
Originality Synthesis-oriented
AI Analysis

This work addresses efficiency issues in image processing for applications like computer vision, but it is incremental as it extends an existing filter method.

The paper tackled the slow computation of the bilateral filter for vector-valued images by developing a fast algorithm using raised-cosine approximations and Monte Carlo sampling, achieving demonstrated speedup over direct implementation in simulations on color images.

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.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes