DocumentCode :
3426338
Title :
Potts Model, Parametric Maxflow and K-Submodular Functions
Author :
Gridchyn, Igor ; Kolmogorov, Vladimir
Author_Institution :
IST Austria, Klosterneuburg, Austria
fYear :
2013
fDate :
1-8 Dec. 2013
Firstpage :
2320
Lastpage :
2327
Abstract :
The problem of minimizing the Potts energy function frequently occurs in computer vision applications. One way to tackle this NP-hard problem was proposed by Kovtun [19, 20]. It identifies a part of an optimal solution by running k maxflow computations, where k is the number of labels. The number of "labeled" pixels can be significant in some applications, e.g. 50-93% in our tests for stereo. We show how to reduce the runtime to O(log k) maxflow computations (or one parametric maxflow computation). Furthermore, the output of our algorithm allows to speed-up the subsequent alpha expansion for the unlabeled part, or can be used as it is for time-critical applications. To derive our technique, we generalize the algorithm of Felzenszwalb et al. [7] for Tree Metrics. We also show a connection to k-sub modular functions from combinatorial optimization, and discuss k-sub modular relaxations for general energy functions.
Keywords :
Potts model; computational complexity; computer vision; minimisation; stereo image processing; trees (mathematics); NP-hard problem; O(log k) maxflow computation; Potts energy function minimization problem; Potts model; combinatorial optimization; computer vision application; k-maxflow computation; k-submodular functions; k-submodular relaxation; labeled pixel number; parametric maxflow computation; stereo; subsequent alpha expansion; tree metrics; Approximation algorithms; Computer vision; Labeling; Measurement; Minimization; NP-hard problem; Recycling;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision (ICCV), 2013 IEEE International Conference on
Conference_Location :
Sydney, NSW
ISSN :
1550-5499
Type :
conf
DOI :
10.1109/ICCV.2013.288
Filename :
6751399
Link To Document :
بازگشت