Combining Support Vector Machines and Markov Random Fields in an Integrated Framework for Contextual Image Classification.
Gabriele MoserSebastiano B. SerpicoPublished in: IEEE Trans. Geosci. Remote. Sens. (2013)
Keyphrases
- markov random field
- image classification
- support vector
- learning machines
- belief propagation
- maximum a posteriori
- higher order
- graph cuts
- pairwise
- random fields
- image restoration
- parameter estimation
- mrf model
- energy function
- potential functions
- conditional random fields
- energy minimization
- potts model
- bag of words
- image segmentation
- textured images
- low level vision
- map inference
- multi label
- map estimation
- message passing
- piecewise constant functions
- support vector machine
- image features
- cross validation
- efficient inference
- feature extraction
- hyperplane
- feature selection
- maximum margin
- information extraction
- denoising
- maximum likelihood
- multiscale
- markov networks
- image processing
- iterative conditional
- image restoration and reconstruction
- cellular automata