Mixed-State Markov Random Fields for Motion Texture Modeling and Segmentation.
Tomás CrivelliBruno Cernuschi-FríasPatrick BouthemyJian-Feng YaoPublished in: ICIP (2006)
Keyphrases
- markov random field
- image segmentation
- energy function
- mrf model
- low level vision
- graph cuts
- textured images
- iterated conditional modes
- random fields
- scene labeling
- energy minimization
- belief propagation
- parameter estimation
- shape prior
- higher order
- unsupervised segmentation
- pairwise
- prior model
- segmentation algorithm
- segmentation method
- potts model
- level set
- mrf models
- min cut
- image restoration
- maximum a posteriori
- discriminative random fields
- multiscale
- object segmentation
- label field
- motion model
- potential functions
- map inference
- conditional random fields
- optical flow estimation
- medical images
- least squares
- image analysis
- iterative conditional
- motion field
- loopy belief propagation
- map estimation
- texture segmentation
- image processing
- bayesian estimation
- markov networks
- computer vision
- image reconstruction
- maximum likelihood