Markov Random Field modeling, inference & learning in computer vision & image understanding: A survey

Abstract

In this paper, we present a comprehensive survey of Markov Random Fields (MRFs) in computer vision and image understanding, with respect to the modeling, the inference and the learning. While MRFs were introduced into the computer vision eld about two decades ago, they started to become a ubiquitous tool for solving visual perception problems around the turn of the millennium following the emergence of efficient inference methods. During the past decade, a variety of MRF models as well as inference and learning methods have been developed for addressing numerous low, mid and high-level vision problems. While most of the literature concerns pairwise MRFs, in recent years we have also witnessed signi cant progress in higher-order MRFs, which substantially enhances the expressiveness of graph-based models and expands the domain of solvable problems. This survey provides a compact and informative summary of the major literature in this research topic.

VPRS-Based Regional Decision Fusion of CNN and MRF Classifications for Very Fine Resolution Remotely Sensed Images

“…The complete correctness of the MLP-MRF results at the non-positive regions are not guaranteed, but one thing is certain: the corresponding MLP-MRF results are much more accurate than those of the CNN. In fact, while the CNN accurately classifies the interiors of objects with spatial feature representations, the MLP-MRF could provide a smooth, but also crisp boundary segmentation with high fidelity [56]. These supplementary characteristics inherent in the MLP-MRF and CNN, are captured well by the proposed VPRS-based MRF-CNN regional decision fusion approach.…”

“…That is, the MLP-MRF depends primarily on the spectral feature differentiation from the MLP with consideration of its spatial connectivity/smoothness [56]. Such characteristics result in similar classification performance to the result of MLP but with less salt and pepper effect.…”

Efficient Bayesian Communication Approach for Smart Agriculture Applications

“…Each node represents a random variable, and the edges represent probabilistic relationships among variables. Models which are comprised of directed edges are known as Bayesian networks, whilst models that are composed of undirected edges are known as Markov Random Fields (MRF) [12]. In this paper, we present an inference approach under the hypothesis of MRF, modeled by means of Factor Graphs.…”

(Hyper)-graphical models in biomedical image analysis

“…Graph-based representations were introduced in computer vision at mid-eighties [2] through Markov Random Fields as a novel mathematical modeling framework constrained though from the lack of efficient inference methods as well as processing power -and became again popular during the past two decades thanks to the development of efficient optimization algorithms [3,4].…”