pROST : A Smoothed Lp-norm Robust Online Subspace Tracking Method for Realtime Background Subtraction in Video
Department of Informatics, Technische Universitat Munchen, Boltzmannstr. 3, 85748 Garching, Germany
arXiv:1302.2073 [cs.CV], (8 Feb 2013)
@article{2013arXiv1302.2073S,
author={Seidel}, F. and {Hage}, C. and {Kleinsteuber}, M.},
title={"{pROST : A Smoothed Lp-norm Robust Online Subspace Tracking Method for Realtime Background Subtraction in Video}"},
journal={ArXiv e-prints},
archivePrefix={"arXiv"},
eprint={1302.2073},
primaryClass={"cs.CV"},
keywords={Computer Science – Computer Vision and Pattern Recognition},
year={2013},
month={feb},
adsurl={http://adsabs.harvard.edu/abs/2013arXiv1302.2073S},
adsnote={Provided by the SAO/NASA Astrophysics Data System}
}
An increasing number of methods for background subtraction use Robust PCA to identify sparse foreground objects. While many algorithms use the L1-norm as a convex relaxation of the ideal sparsifying function, we approach the problem with a smoothed Lp-norm and present pROST, a method for robust online subspace tracking. The algorithm is based on alternating minimization on manifolds. Implemented on a graphics processing unit it achieves realtime performance. Experimental results on a state-of-the-art benchmark for background subtraction on real-world video data indicate that the method succeeds at a broad variety of background subtraction scenarios, and it outperforms competing approaches when video quality is deteriorated by camera jitter.
February 15, 2013 by hgpu
