Tree-based iterated local search for Markov random fields with applications in image analysis

TL;DR

This paper introduces Tree-based Iterated Local Search (T-ILS), a novel method leveraging tree structures within Markov random fields to efficiently solve MAP assignment problems in image analysis, showing competitive results with reduced computation.

Contribution

The paper presents a new tree-based ILS approach that efficiently explores large neighborhoods in MRFs without requiring specific cost function properties.

Findings

T-ILS achieves significant computational gains.

T-ILS performs competitively in stereo matching and denoising.

Method is effective on both simulated and real-world data.

Abstract

The \emph{maximum a posteriori} (MAP) assignment for general structure Markov random fields (MRFs) is computationally intractable. In this paper, we exploit tree-based methods to efficiently address this problem. Our novel method, named Tree-based Iterated Local Search (T-ILS) takes advantage of the tractability of tree-structures embedded within MRFs to derive strong local search in an ILS framework. The method efficiently explores exponentially large neighborhood and does so with limited memory without any requirement on the cost functions. We evaluate the T-ILS in a simulation of Ising model and two real-world problems in computer vision: stereo matching, image denoising. Experimental results demonstrate that our methods are competitive against state-of-the-art rivals with a significant computational gain.