Minimum message length estimation of mixtures of multivariate Gaussian and von Mises-Fisher distributions.
Parthan KasarapuLloyd AllisonPublished in: CoRR (2015)
Keyphrases
- multivariate gaussian
- maximum likelihood estimation
- minimum message length
- mixture model
- von mises fisher
- expectation maximization
- em algorithm
- finite mixtures
- gaussian distribution
- maximum likelihood
- gaussian mixture model
- parameter estimation
- unsupervised learning
- density estimation
- multi variate
- probabilistic model
- gaussian mixture
- generative model
- noise model
- covariance matrix
- probability density function
- probability density
- unit sphere
- mixture components
- image segmentation
- bayesian framework
- probability distribution
- computer vision
- medical images
- semi supervised
- objective function