Support vector driven Markov random fields towards DTI segmentation of the human skeletal muscle.
Radhouène NejiGilles FleuryJean-François DeuxAlain RahmouniGuillaume BassezAlexandre VignaudNikos ParagiosPublished in: ISBI (2008)
Keyphrases
- markov random field
- energy function
- image segmentation
- mrf model
- low level vision
- support vector
- textured images
- iterated conditional modes
- label field
- graph cuts
- random fields
- scene labeling
- mrf models
- higher order
- segmentation algorithm
- belief propagation
- energy minimization
- pairwise
- maximum a posteriori
- parameter estimation
- image restoration
- segmentation method
- prior model
- conditional random fields
- unsupervised segmentation
- discriminative random fields
- level set
- shape prior
- min cut
- image analysis
- multiscale
- color images
- potential functions
- feature selection
- map inference
- map estimation
- potts model
- region growing
- image labeling
- diffusion tensor imaging
- loopy belief propagation
- semantic segmentation
- image reconstruction
- image processing