An incremental principal component pursuit algorithm via projections onto the ℓ1 ball

Paul Rodriguez, Brendt Wohlberg

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Scopus citations

Abstract

Video background modeling, used to detect moving objects in digital videos, is a ubiquitous pre-processing step in computer vision applications. Principal Component Pursuit (PCP) PCP is among the leading methods for this problem. In this paper we proposed a new convex formulation for PCP, substituting the standard ℓ1 regularization with a projection onto the ℓ1-ball. This formulation offers an advantage over the known incremental PCP methods in practical parameter selection and ghosting suppression, while retaining the ability to be implemented in a fully incremental fashion, keeping all the desired properties related to such PCP methods (low memory footprint, adaptation to changes in the background, computational complexity that allows online processing).

Original languageEnglish
Title of host publicationProceedings of the 2017 IEEE 24th International Congress on Electronics, Electrical Engineering and Computing, INTERCON 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509063628
DOIs
StatePublished - 20 Oct 2017
Event24th IEEE International Congress on Electronics, Electrical Engineering and Computing, INTERCON 2017 - Cusco, Peru
Duration: 15 Aug 201718 Aug 2017

Publication series

NameProceedings of the 2017 IEEE 24th International Congress on Electronics, Electrical Engineering and Computing, INTERCON 2017

Conference

Conference24th IEEE International Congress on Electronics, Electrical Engineering and Computing, INTERCON 2017
Country/TerritoryPeru
CityCusco
Period15/08/1718/08/17

Keywords

  • Incremental Principal Component Pursuit
  • Video Background Modeling
  • ℓ Ball projection

Fingerprint

Dive into the research topics of 'An incremental principal component pursuit algorithm via projections onto the ℓ1 ball'. Together they form a unique fingerprint.

Cite this