Forming A Random Field via Stochastic Cliques: From Random Graphs to Fully Connected Random Fields

TL;DR

This paper introduces a stochastic clique-based framework for fully-connected random fields that enables efficient structured inference in image segmentation, overcoming computational challenges associated with long-range interactions.

Contribution

It proposes a novel stochastic clique approach that allows fully-connected random fields to be computationally feasible without restricting potential functions.

Findings

Competitive segmentation performance compared to existing methods

Efficient inference using graph cuts

Addresses short-boundary bias in local models

Abstract

Random fields have remained a topic of great interest over past decades for the purpose of structured inference, especially for problems such as image segmentation. The local nodal interactions commonly used in such models often suffer the short-boundary bias problem, which are tackled primarily through the incorporation of long-range nodal interactions. However, the issue of computational tractability becomes a significant issue when incorporating such long-range nodal interactions, particularly when a large number of long-range nodal interactions (e.g., fully-connected random fields) are modeled. In this work, we introduce a generalized random field framework based around the concept of stochastic cliques, which addresses the issue of computational tractability when using fully-connected random fields by stochastically forming a sparse representation of the random field. The proposed framework allows for efficient structured inference using fully-connected random fields without any restrictions on the potential functions that can be utilized. Several realizations of the proposed framework using graph cuts are presented and evaluated, and experimental results demonstrate that the proposed framework can provide competitive performance for the purpose of image segmentation when compared to existing fully-connected and principled deep random field frameworks.