Local Principal Component Pursuit for Nonlinear Datasets

Abstract

A robust version of Principal Component Analysis (PCA) can be constructed via a decomposition of a data matrix into low-rank and sparse components, the former representing a low-dimensional linear model of the data, and the latter representing sparse deviations from the low-dimensional subspace. This decomposition has been shown to be highly effective, but the underlying model is not appropriate when the data are not modeled well by a single low-dimensional subspace. We construct a new decomposition corresponding to a more general underlying model consisting of a union of low-dimensional subspaces, and demonstrate its performance on a video background removal problem.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2012
Accession Number
ADA568726

Entities

People

  • Brendt Wohlberg
  • James Theiler
  • Rick Chartrand

Organizations

  • Los Alamos National Laboratory

Tags

Communities of Interest

  • Air Platforms

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Compressed Sensing
  • Dictionaries
  • Factor Analysis
  • Index Terms
  • Information Operations
  • Linear Systems
  • Mathematics
  • Sequences
  • Signal Processing
  • Standards
  • Three Dimensional

Readers

  • Approximation Theory.
  • Neural Network Machine Learning.