Dimensionality reduction and principal surfaces via Kernel Map Manifolds.
Samuel GerberTolga TasdizenRoss T. WhitakerPublished in: ICCV (2009)
Keyphrases
- dimensionality reduction
- laplacian eigenmaps
- manifold learning
- low dimensional
- feature space
- kernel pca
- high dimensional
- nonlinear dimensionality reduction
- kernel learning
- high dimensional data
- input space
- kernel trick
- principal curves
- principal component analysis
- class separability
- high dimensionality
- data representation
- low dimensional spaces
- embedding space
- graph embedding
- kernel methods
- principal components
- manifold structure
- kernel discriminant analysis
- kernel function
- dimensionality reduction methods
- pattern recognition
- feature extraction
- neighborhood preserving
- latent space
- free form
- maximum a posteriori
- data points
- structure preserving
- geodesic paths
- feature selection
- kernel matrix
- range data
- euclidean space
- intrinsic dimensionality
- riemannian manifolds
- lower dimensional
- diffusion maps
- underlying manifold
- pattern recognition and machine learning
- linear discriminant analysis
- neural network
- feature maps
- spectral clustering
- nonlinear manifold