A Geometrical Study of Matching Pursuit Parametrization

DOI10.1109/TSP.2008.917379zbMATH Open1390.94227arXiv0801.3372OpenAlexW2057867606MaRDI QIDQ4568940FDOQ4568940

Authors: Laurent Jacques, Christophe De Vleeschouwer 

Publication date: 27 June 2018

Published in: IEEE Transactions on Signal Processing (Search for Journal in Brave)

Abstract: This paper studies the effect of discretizing the parametrization of a dictionary used for Matching Pursuit decompositions of signals. Our approach relies on viewing the continuously parametrized dictionary as an embedded manifold in the signal space on which the tools of differential (Riemannian) geometry can be applied. The main contribution of this paper is twofold. First, we prove that if a discrete dictionary reaches a minimal density criterion, then the corresponding discrete MP (dMP) is equivalent in terms of convergence to a weakened hypothetical continuous MP. Interestingly, the corresponding weakness factor depends on a density measure of the discrete dictionary. Second, we show that the insertion of a simple geometric gradient ascent optimization on the atom dMP selection maintains the previous comparison but with a weakness factor at least two times closer to unity than without optimization. Finally, we present numerical experiments confirming our theoretical predictions for decomposition of signals and images on regular discretizations of dictionary parametrizations.

Full work available at URL: https://arxiv.org/abs/0801.3372

Recommendations

• The Geometry of Compressed Sensing
• The Geometry of Sparse Analysis Regularization
• A geometric analysis of phase retrieval
• A matching pursuit for the sphere based on the Poisson semigroup
• A geometric framework for sparse matrix problems
• A geometric approach to the singular filtering problem
• A geometrical stability condition for compressed sensing
• Publication:4893182
• Geometric Methods in Signal and Image Analysis

Mathematics Subject Classification ID

Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Cited In (6)

• The sliding Frank-Wolfe algorithm and its application to super-resolution microscopy
• Spectral compressive sensing
• Geometry of the matching distance for 2D filtering functions
• MultiDimensional Sparse Super-Resolution
• Geometric Video Approximation Using Weighted Matching Pursuit
• The Geometry of Sparse Analysis Regularization

This page was built for publication: A Geometrical Study of Matching Pursuit Parametrization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4568940)