Revisiting the Linear Programming Relaxation Approach to Gibbs Energy Minimization and Weighted Constraint Satisfaction.
Tomás WernerPublished in: IEEE Trans. Pattern Anal. Mach. Intell. (2010)
Keyphrases
- object recognition
- weighted constraint satisfaction
- energy minimization
- markov random field
- energy function
- graph cuts
- problems in computer vision
- maximum a posteriori
- belief propagation
- image segmentation
- low level vision
- pairwise
- global minimum
- parameter estimation
- higher order
- interactive segmentation
- image restoration
- curve evolution
- min cut
- max flow
- global minimization
- constraint satisfaction
- image processing
- conditional random fields
- dual decomposition
- max flow min cut
- dynamic programming
- message passing
- markov chain monte carlo
- simulated annealing