Fisher-Rao and pullback Hilbert cone distances on the multivariate Gaussian manifold with applications to simplification and quantization of mixtures.
Frank NielsenPublished in: TAG-ML (2023)
Keyphrases
- multivariate gaussian
- maximum likelihood estimation
- covariance matrix
- geodesic distance
- gaussian distribution
- expectation maximization
- distance measure
- intravascular ultrasound
- mixture model
- em algorithm
- multiresolution
- euclidean distance
- distance function
- low dimensional
- manifold learning
- euclidean space
- image space
- maximum likelihood
- gaussian mixture model
- probability distribution
- sample size
- image segmentation
- computational complexity
- similarity measure