Riemannian Generalized Gaussian Distributions on the Space of SPD Matrices for Image Classification.
Zakariae AbbadAhmed Drissi El MalianiMohammed El HassouniMohamed Tahar Kadaoui AbbassiLionel BombrunYannick BerthoumieuPublished in: IEEE Access (2024)
Keyphrases
- spd matrices
- gaussian distribution
- symmetric positive definite
- riemannian manifolds
- image classification
- lie group
- covariance matrices
- log euclidean
- euclidean space
- positive definite
- finite dimensional
- distribution function
- maximum likelihood
- vector space
- kl divergence
- geodesic distance
- diffusion tensor
- covariance matrix
- riemannian metric
- gaussian mixture model
- expectation maximization
- parameter space
- infinite dimensional
- feature extraction
- vector field
- von neumann
- higher dimensional
- kernel matrix
- image segmentation
- mean shift
- face recognition