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

被引:2
作者
Truyen Tran [1 ,2 ]
Dinh Phung [2 ]
Venkatesh, Svetha [2 ]
机构
[1] Curtin Univ, Dept Comp, Bentley, WA 6102, Australia
[2] Deakin Univ, Ctr Pattern Recognit & Data Analyt, Waurn Ponds, Vic 3216, Australia
关键词
Iterated local search; Strong local search; Belief propagation; Markov random fields; MAP assignment; BELIEF-PROPAGATION; STATISTICAL-ANALYSIS; ENERGY MINIMIZATION; MAP ESTIMATION; SEGMENTATION; MRF; DISTRIBUTIONS; OPTIMIZATION; RELAXATIONS; ALGORITHM;
D O I
10.1007/s10732-014-9270-1
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The maximum a posteriori assignment for general structure Markov random fields 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 neighborhoods using a limited memory without any requirement on the cost functions. We evaluate the T-ILS on a simulated Ising model and two real-world vision problems: stereo matching and image denoising. Experimental results demonstrate that our methods are competitive against state-of-the-art rivals with significant computational gain.
引用
收藏
页码:25 / 45
页数:21
相关论文
共 49 条
[1]   A survey of very large-scale neighborhood search techniques [J].
Ahuja, RK ;
Ergun, Ö ;
Orlin, JB ;
Punnen, AP .
DISCRETE APPLIED MATHEMATICS, 2002, 123 (1-3) :75-102
[2]  
[Anonymous], 2006, Proceedings of the 23rd International Conference on Machine Learning (ICML 2006), DOI DOI 10.1145/1143844.1143937
[3]  
[Anonymous], MARKOV FIELDS UNPUB
[4]  
BESAG J, 1974, J ROY STAT SOC B MET, V36, P192
[5]  
BESAG J, 1986, J R STAT SOC B, V48, P259
[6]  
Biba M, 2008, LECT NOTES ARTIF INT, V5194, P59, DOI 10.1007/978-3-540-85928-4_9
[7]   Fast approximate energy minimization via graph cuts [J].
Boykov, Y ;
Veksler, O ;
Zabih, R .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2001, 23 (11) :1222-1239
[8]  
Brown D.F., 2002, ARTIFICIAL EVOLUTION, P35
[9]   APPROXIMATING DISCRETE PROBABILITY DISTRIBUTIONS WITH DEPENDENCE TREES [J].
CHOW, CK ;
LIU, CN .
IEEE TRANSACTIONS ON INFORMATION THEORY, 1968, 14 (03) :462-+
[10]   Image registration with iterated local search [J].
Cordón, O ;
Damas, S .
JOURNAL OF HEURISTICS, 2006, 12 (1-2) :73-94