Non-parametric Higher-Order Random Fields for Image Segmentation.
Pablo Márquez-NeilaPushmeet KohliCarsten RotherLuis BaumelaPublished in: ECCV (6) (2014)
Keyphrases
- random fields
- markov random field
- higher order
- image segmentation
- conditional random fields
- graph cuts
- markov random field model
- mrf model
- maximum a posteriori
- pairwise
- parameter estimation
- random field models
- image restoration
- textured images
- belief propagation
- energy function
- autoregressive
- natural images
- potential functions
- pseudo likelihood
- gibbs sampler
- energy minimization
- nonparametric density estimation
- random field model
- markov models
- unsupervised image segmentation
- object segmentation
- maximum entropy
- non stationary
- unsupervised segmentation
- region growing
- texture model
- contour detection
- shape prior
- probability density function
- active contours
- hidden markov models
- multiscale