Multi-temperature annealing: a new approach for the energy-minimization of hierarchical Markov random field models.
Josiane ZerubiaZoltan KatoMarc BerthodPublished in: ICPR (1) (1994)
Keyphrases
- energy minimization
- markov random field models
- image segmentation
- energy function
- graph cuts
- markov random field
- image restoration
- problems in computer vision
- global minimum
- belief propagation
- low level vision
- max flow min cut
- simulated annealing
- dual decomposition
- interactive segmentation
- image analysis
- image dependent
- max flow
- level set
- min cut
- global minimization
- multiscale
- lp relaxation
- dynamic programming
- deformable template
- higher order
- curve evolution
- computer vision
- object recognition