Riemannian joint dimensionality reduction and dictionary learning on symmetric positive definite manifolds.
Hiroyuki KasaiBamdev MishraPublished in: EUSIPCO (2018)
Keyphrases
- dictionary learning
- symmetric positive definite
- dimensionality reduction
- riemannian manifolds
- sparse representation
- manifold learning
- low dimensional
- sparse coding
- euclidean space
- feature space
- high dimensional data
- high dimensional
- distribution function
- euclidean distance
- data points
- feature extraction
- unsupervised learning
- principal component analysis
- positive definite
- pattern recognition
- mean shift
- geodesic distance
- kernel pca
- diffusion tensor
- kernel matrix
- lie group
- metric learning
- feature selection
- multi class classification
- parameter space
- image patches
- vector space
- geometric structure
- reproducing kernel hilbert space
- image classification
- face recognition
- computer vision
- natural images
- nearest neighbor