Kernel Methods on Riemannian Manifolds with Gaussian RBF Kernels

Abstract

In this paper, we develop an approach to exploiting kernel methods with manifold-valued data. In many computer vision problems, the data can be naturally represented as points on a Riemannian manifold. Due to the non-Euclidean geometry of Riemannian manifolds, usual Euclidean computer vision and machine learning algorithms yield inferior results on such data. In this paper, we define Gaussian radial basis function (RBF)-based positive definite kernels on manifolds that permit us to embed a given manifold with a corresponding metric in a high dimensional reproducing kernel Hilbert space. These kernels make it possible to utilize algorithms developed for linear spaces on nonlinear manifold-valued data. Since the Gaussian RBF defined with any given metric is not always positive definite, we present a unified framework for analyzing the positive definiteness of the Gaussian RBF on a generic metric space. We then use the proposed framework to identify positive definite kernels on two specific manifolds commonly encountered in computer vision: the Riemannian manifold of symmetric positive definite matrices and the Grassmann manifold, i.e., the Riemannian manifold of linear subspaces of a Euclidean space. We show that many popular algorithms designed for Euclidean spaces, such as support vector machines, discriminant analysis and principal component analysis can be generalized to Riemannian manifolds with the help of such positive definite Gaussian kernels.

References in corpus (2)

• Non-Euclidean statistics for covariance matrices, with applications to diffusion tensor imaging
• Sparse Coding and Dictionary Learning for Symmetric Positive Definite Matrices: A Kernel Approach

Cited by in corpus (29)

• Person Re-identification in Aerial Imagery
• Riemannian stochastic variance reduced gradient algorithm with retraction and vector transport
• Kernel clustering: density biases and solutions
• Dimensionality reduction based on Distance Preservation to Local Mean (DPLM) for SPD matrices and its application in BCI
• Deep Covariance Descriptors for Facial Expression Recognition
• Bayesian Optimization Meets Riemannian Manifolds in Robot Learning
• A Novel Space-Time Representation on the Positive Semidefinite Con for Facial Expression Recognition
• Topology of surface electromyogram signals: hand gesture decoding on Riemannian manifolds
• Riemannian stochastic variance reduced gradient on Grassmann manifold
• Nonparametric Continuous Sensor Registration
• Cross Euclidean-to-Riemannian Metric Learning with Application to Face Recognition from Video
• Leveraging tensor kernels to reduce objective function mismatch in deep clustering
• R-Mixup: Riemannian Mixup for Biological Networks
• Gaussian kernels on non-simply-connected closed Riemannian manifolds are never positive definite
• Emotion Recognition by Body Movement Representation on the Manifold of Symmetric Positive Definite Matrices
• Differential geometry with extreme eigenvalues in the positive semidefinite cone
• Riemannian stochastic recursive momentum method for non-convex optimization
• Accelerated kernel discriminant analysis
• High-Dimensional Bayesian Optimization via Nested Riemannian Manifolds
• Interpretable discriminant analysis for functional data supported on random nonlinear domains with an application to Alzheimer's disease
• Marginalising over Stationary Kernels with Bayesian Quadrature
• Locality Preserving Projections for Grassmann manifold
• Image segmentation with superpixel-based covariance descriptors in low-rank representation
• EEG Representation Using Multi-instance Framework on The Manifold of Symmetric Positive Definite Matrices for EEG-based Computer Aided Diagnosis
• Adversarially robust clustering with optimality guarantees
• Optimized Kernel-based Projection Space of Riemannian Manifolds
• Volume entropy and information flow in a brain graph
• Probabilistic Learning Vector Quantization on Manifold of Symmetric Positive Definite Matrices
• Alternating Diffusion Map Based Fusion of Multimodal Brain Connectivity Networks for IQ Prediction