Optimization of Discrete Markov Random Fields via Dual Decomposition

نویسندگان

  • Nikos Komodakis
  • Nikos Paragios
  • Georgios Tziritas
چکیده

A new message-passing scheme for MRF optimization is proposed in this paper. This scheme inherits better theoretical properties than all other state-of-the-art message passing methods and in practice performs equally well/outperforms them. It is based on the very powerful technique of Dual Decomposition [1] and leads to an elegant and general framework for understanding/designing message-passing algorithms that can provide new insights into existing techniques. Promising experimental results and comparisons with the state of the art demonstrate the extreme theoretical and practical potentials of our approach. Key-words: Markov Random Fields, Dual Decomposition, Linear Programming, Optimization. ∗ This is a pending submission to the IEEE International Conference on Computer Vision 2007 (ICCV’07). † [email protected], [email protected] Ecole Centrale de Paris, University of Crete ‡ [email protected] Ecole Centrale de Paris § [email protected] University of Crete Optimisation de Champs de Markov via Décomposition Duale Résumé : Dans cet article, nous proposons un nouveau schéma de propagation de messages pour l’optimisation de champs de Markov. Ce schéma hérite de meilleures propriétés théoriques que toute autre méthode de propagation de messages décrite dans l’état de l’art. En pratique, notre schéma a au moins d’aussi bon résultats que ces méthodes, sinon meilleurs. Il est basé sur la technique puissante de la décomposition duale et mène à un cadre élégant et général pour la compréhension et la conception d’algorithmes de propagation de messages qui peuvent fournir de nouvelles perspectives pour les techniques existantes. Les résultats expérimentaux prometteurs et les comparaisons avec l’état de l’art démontrent le potentiel théorique et pratique extrême de notre approche. Mots-clés : Champs de Markov, Décomposition duale, Programmation linéaire, Optimisation. Optimization of Discrete Markov Random Fields via Dual Decomposition 3

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Partial Optimality of Dual Decomposition for MAP Inference in Pairwise MRFs

Markov random fields (MRFs) are a powerful tool for modelling statistical dependencies for a set of random variables using a graphical representation. An important computational problem related to MRFs, called maximum a posteriori (MAP) inference, is finding a joint variable assignment with the maximal probability. It is well known that the two popular optimisation techniques for this task, lin...

متن کامل

Constrained Discrete Optimization via Dual Space Search

This paper proposes a novel algorithm for solving NP-hard constrained discrete minimization problems whose unconstrained versions are solvable in polynomial time such as constrained submodular function minimization. Applications of our algorithm include constrained MAP inference in Markov Random Fields, and energy minimization in various computer vision problems. Our algorithm assumes the exist...

متن کامل

Continuous Relaxation of MAP Inference: A Nonconvex Perspective

In this paper, we study a nonconvex continuous relaxation of MAP inference in discrete Markov random fields (MRFs). We show that for arbitrary MRFs, this relaxation is tight, and a discrete stationary point of it can be easily reached by a simple block coordinate descent algorithm. In addition, we study the resolution of this relaxation using popular gradient methods, and further propose a more...

متن کامل

Distributed Non-Convex ADMM-inference in Large-scale Random Fields

We propose a parallel and distributed algorithm for solving discrete labeling problems in large scale random fields. Our approach is motivated by the following observations: i) very large scale image and video processing problems, such as labeling dozens of million pixels with thousands of labels, are routinely faced in many application domains; ii) the computational complexity of the current s...

متن کامل

Efficient Inference of Continuous Markov Random Fields with Polynomial Potentials

In this paper, we prove that every multivariate polynomial with even degree can be decomposed into a sum of convex and concave polynomials. Motivated by this property, we exploit the concave-convex procedure to perform inference on continuous Markov random fields with polynomial potentials. In particular, we show that the concave-convex decomposition of polynomials can be expressed as a sum-of-...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007