A Bayesian approach to geometric subspace estimation.

DOI10.1109/78.839985MaRDI QIDQ2734333zbMATH OpenOpenAlexFDO

Authors Anuj Strivastava

Publication date 2000

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

Full work available at URL https://semanticscholar.org/paper/84d849d34c677632f073f6ea3848a89dbe9cf9ab

zbMATH Keywords

Grassman manifoldMonte Carlo samplingsubspace estimation

Mathematics Subject Classification ID

Discrete event control/observation systems (93C65) Identification in stochastic control theory (93E12) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Review text

The matter of subspace estimation is of considerable interest in engineering problems. Here the author presents a Bayesian framework for subspace estimation and tracking. Such subspaces are represented by their unique complex projection matrices, these being viewed as elements of a complex Grassman manifold. A posterior density is derived for the elements of the unitary group \(U(n)\), and pertinent expectations are empirically generated.

Recommendations

• Bayesian and geometric subspace tracking
• Geometry and multiple direction estimation
• Invariance of subspace based estimators
• Signal space geometry
• Statistical estimation of a subspace in a complex space

Cited in

(6)

• Geometry and multiple direction estimation
• Jump-diffusion Markov processes on orthogonal groups for object pose estimation
• Bayesian point estimation of the cointegration space
• Welcome to Riemannian computing in computer vision
• Bayesian and geometric subspace tracking
• Local convergence of an algorithm for subspace identification from partial data

This page was built for publication: A Bayesian approach to geometric subspace estimation.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2734333)