Global to push GA events into
skip to main content

Title: Using dynamic mode decomposition for real-time background/foreground separation in video

The technique of dynamic mode decomposition (DMD) is disclosed herein for the purpose of robustly separating video frames into background (low-rank) and foreground (sparse) components in real-time. Foreground/background separation is achieved at the computational cost of just one singular value decomposition (SVD) and one linear equation solve, thus producing results orders of magnitude faster than robust principal component analysis (RPCA). Additional techniques, including techniques for analyzing the video for multi-resolution time-scale components, and techniques for reusing computations to allow processing of streaming video in real time, are also described herein.
Inventors:
; ; ; ;
Issue Date:
OSTI Identifier:
1361416
Assignee:
UNIVERSITY OF WASHINGTON DOEEE
Patent Number(s):
9,674,406
Application Number:
14/828,396
Contract Number:
EE0006785; FA9550-09-0174
Resource Relation:
Patent File Date: 2015 Aug 17
Research Org:
Univ. of Washington, Seattle, WA (United States)
Sponsoring Org:
USDOE
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING

Works referenced in this record:

Comparative study of background subtraction algorithms
journal, July 2010
  • Benezeth, Y.; Jodoin, Pierre-Marc; Emile, Bruno
  • Journal of Electronic Imaging, Vol. 19, Issue 3, Article No. 033003
  • DOI: 10.1117/1.3456695

Variants of Dynamic Mode Decomposition: Boundary Condition, Koopman, and Fourier Analyses
journal, April 2012
  • Chen, Kevin K.; Tu, Jonathan H.; Rowley, Clarence W.
  • Journal of Nonlinear Science, Vol. 22, Issue 6, p. 887-915
  • DOI: 10.1007/s00332-012-9130-9

A Framework for Robust Subspace Learning
journal, January 2003
  • De la Torre, Fernando; Black, Michael J.
  • International Journal of Computer Vision, Vol. 54, Issue 1-3, p. 117-142
  • DOI: 10.1023/A:1023709501986

Applications of the dynamic mode decomposition
journal, August 2010
  • Schmid, P. J.; Li, L.; Juniper, M. P.
  • Theoretical and Computational Fluid Dynamics, Vol. 25, Issue 1-4, p. 249-259
  • DOI: 10.1007/s00162-010-0203-9

Fusion of Multiple Tracking Algorithms for Robust People Tracking
book, January 2002
  • Siebel, Nils T.; Maybank, Steve; Goos, Gerhard
  • Computer Vision — ECCV 2002, p. 373-387
  • DOI: 10.1007/3-540-47979-1_25