Fast Polynomial Approximation of Heat Kernel Convolution on Manifolds and Its Application to Brain Sulcal and Gyral Graph Pattern Analysis.
Shih-Gu HuangIlwoo LyuAnqi QiuMoo K. ChungPublished in: IEEE Trans. Medical Imaging (2020)
Keyphrases
- pattern analysis
- cortical surface
- heat kernel
- polynomial approximation
- graph laplacian
- pointwise
- graph structure
- euclidean space
- laplacian eigenmaps
- shape analysis
- human brain
- pattern recognition
- manifold structure
- weight matrix
- geodesic distance
- minimum spanning tree
- euclidean distance
- random walk
- laplace beltrami
- shape descriptors
- weighted sum
- spanning tree
- cortical thickness
- spectral analysis
- image analysis
- pattern classification
- manifold learning
- scale space
- lower bound
- spectral clustering
- feature extraction
- low dimensional
- active shape model
- graph theory
- computational intelligence
- directed graph
- weighted graph
- neural network
- riemannian manifolds
- image processing
- data analysis
- high dimensional
- kernel machines
- neighborhood graph
- multiscale
- parameter space
- vector space
- face recognition
- graphical models