Unsupervised co-segmentation based on a new global GMM constraint in MRF.
Hongkai YuMin XianXiaojun QiPublished in: ICIP (2014)
Keyphrases
- image segmentation
- energy function
- markov random field
- graph cuts
- scene labeling
- gaussian mixture model
- low level vision
- fully unsupervised
- segmentation method
- segmentation algorithm
- energy minimization
- segmentation of natural images
- bayesian image segmentation
- text segmentation
- level set
- background subtraction
- medical images
- shape prior
- segmentation of textured images
- mrf model
- smoothness constraint
- feature vectors
- image analysis
- gaussian mixture modeling
- supervised learning
- global consistency
- higher order
- outdoor scenes
- constraint networks
- object segmentation
- high order
- multiscale
- semi supervised
- normalized cut
- min cut
- global shape
- word segmentation
- textured images
- em algorithm
- conditional random fields
- expectation maximization