Title :
Tensor Power Method for Efficient MAP Inference in Higher-order MRFs
Author :
Semenovich, Dimitri ; Sowmya, Arcot
Author_Institution :
Univ. of New South Wales, Sydney, NSW, Australia
Abstract :
We present a new efficient algorithm for maximizing energy functions with higher order potentials suitable for MAP inference in discrete MRFs. Initially we relax integer constraints on the problem and obtain potential label assignments using higher-order (tensor) power method. Then we utilise an ascent procedure similar to the classic ICM algorithm to converge to a solution meeting the original integer constraints.
Keywords :
Markov processes; inference mechanisms; maximum likelihood estimation; random processes; tensors; MAP inference; discrete MRF; higher-order tensor power method; integer constraints; tensor power method; Approximation algorithms; Approximation methods; Belief propagation; Inference algorithms; Manganese; Markov processes; Tensile stress; Belief Propagation; Higher Order Power Method; MAP inference; MRF;
Conference_Titel :
Pattern Recognition (ICPR), 2010 20th International Conference on
Conference_Location :
Istanbul
Print_ISBN :
978-1-4244-7542-1
DOI :
10.1109/ICPR.2010.185
