Scalable real-time object recognition and segmentation via cascaded, discriminative Markov random fields.
Paul VernazaDaniel D. LeePublished in: ICRA (2010)
Keyphrases
- markov random field
- object recognition
- discriminative random fields
- low level vision
- image segmentation
- energy function
- mrf model
- textured images
- iterated conditional modes
- graph cuts
- computer vision
- mrf models
- label field
- maximum a posteriori
- higher order
- scene labeling
- parameter estimation
- piecewise constant functions
- conditional random fields
- energy minimization
- segmentation method
- belief propagation
- random fields
- prior model
- image restoration
- pairwise
- unsupervised segmentation
- image understanding
- min cut
- object segmentation
- crf model
- potential functions
- map estimation
- image analysis
- multiscale
- segmentation algorithm
- shape model
- potts model
- scene understanding
- loopy belief propagation
- natural images
- level set
- object detection
- image processing
- denoising
- gibbs energy
- image matching
- texture segmentation