Incorporating a Metropolis method in a distribution estimation using Markov random field algorithm.
Siddhartha ShakyaJohn A. W. McCallDeryck Forsyth BrownPublished in: Congress on Evolutionary Computation (2005)
Keyphrases
- markov random field
- parameter estimation
- energy function
- mrf model
- map estimation
- segmentation algorithm
- dynamic programming
- loopy belief propagation
- k means
- simulated annealing
- prior information
- maximum a posteriori probability
- belief propagation
- probabilistic model
- segmentation method
- prior model
- message passing
- markov chain monte carlo
- em algorithm
- texture model
- pairwise
- image segmentation
- relaxation labeling
- label field
- iterated conditional modes
- similarity measure
- energy minimization
- bayesian framework
- graph cuts
- expectation maximization
- posterior distribution
- test images
- generalized gaussian
- maximum a posterior
- gaussian distribution
- maximum a posteriori
- closed form
- active contours
- labeling problems
- mrf optimization
- edge detection
- prior knowledge