A Cramér-Rao Bound Characterization of the EM-Algorithm Mean Speed of Convergence.
Cédric HerzetValéry RamonAlexandre RenauxLuc VandendorpePublished in: IEEE Trans. Signal Process. (2008)
Keyphrases
- rao bound
- em algorithm
- expectation maximization
- mixture model
- maximum likelihood
- gaussian mixture model
- maximum likelihood estimation
- gaussian mixture
- parameter estimation
- generative model
- hyperparameters
- log likelihood function
- expectation maximisation
- probabilistic model
- probability density function
- maximum a posteriori
- incomplete data
- convergence rate
- object recognition
- mixture modeling
- depth from defocus
- model based clustering
- unsupervised learning
- spatial domain
- input image
- image segmentation
- penalized likelihood