Efficient belief propagation for higher-order cliques using linear constraint nodes.
Brian PotetzTai Sing LeePublished in: Comput. Vis. Image Underst. (2008)
Keyphrases
- higher order cliques
- belief propagation
- markov random field
- graphical models
- higher order
- linear constraints
- message passing
- graph cuts
- approximate inference
- pairwise
- loopy belief propagation
- stereo matching
- energy minimization
- fixed point
- factor graphs
- belief networks
- markov networks
- free energy
- high order
- maximum a posteriori
- energy function
- computer vision
- conditional random fields
- image restoration
- post processing
- model selection
- dynamic programming
- multiscale
- bayesian networks