Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices.
Sadeep JayasumanaRichard I. HartleyMathieu SalzmannHongdong LiMehrtash Tafazzoli HarandiPublished in: CVPR (2013)
Keyphrases
- riemannian manifolds
- kernel methods
- reproducing kernel hilbert space
- feature space
- euclidean space
- kernel function
- kernel pca
- log euclidean
- support vector
- learning problems
- machine learning
- kernel matrix
- vector space
- multi class classification
- geometric structure
- support vector machine
- covariance matrix
- mean shift
- finite dimensional
- parameter space
- manifold learning
- data points
- riemannian metric
- covariance matrices
- kernel machines
- graph kernels
- feature selection
- lie group
- positive definite
- vector valued
- image processing
- geodesic distance