brendt.wohlberg.net
HomePublications
› Publications
› Software

Cite Details

Paul Rodríguez and Brendt Wohlberg, "An Incremental Principal Component Pursuit Algorithm via Projections onto the 1 Ball", in Proceedings of International Conference on Electronics, Electrical Engineering and Computing (INTERCON), (Cusco, Perú), doi:10.1109/INTERCON.2017.8079645, pp. 1--4, Aug 2017

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 1 regularization with a projection onto the 1-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).

BibTeX Entry

@inproceedings{rodriguez-2017-incremental,
author = {Paul Rodr\'{i}guez and Brendt Wohlberg},
title = {An Incremental Principal Component Pursuit Algorithm via Projections onto the $\ell_{1}$ Ball},
year = {2017},
month = Aug,
urlpdf = {http://brendt.wohlberg.net/publications/pdf/rodriguez-2017-incremental.pdf},
booktitle = {Proceedings of International Conference on Electronics, Electrical Engineering and Computing (INTERCON)},
address = {Cusco, Per\'{u}},
doi = {10.1109/INTERCON.2017.8079645},
pages = {1--4}
}