Probabilistic image-based tracking: improving particle filtering

Authors:
Daniel Rowe;Ignasi Rius;Jordi Gonzàlez;Xavier Roca;Juan J. Villanueva
Affiliations:
Computer Vision Centre/Department of Computing Science, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain;Computer Vision Centre/Department of Computing Science, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain;Computer Vision Centre/Department of Computing Science, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain;Computer Vision Centre/Department of Computing Science, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain;Computer Vision Centre/Department of Computing Science, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain
Venue:
IbPRIA'05 Proceedings of the Second Iberian conference on Pattern Recognition and Image Analysis - Volume Part I
Year:
2005

Citing 5
Cited 1

CONDENSATION—Conditional Density Propagation forVisual Tracking

International Journal of Computer Vision
How Does CONDENSATION Behave with a Finite Number of Samples?

ECCV '00 Proceedings of the 6th European Conference on Computer Vision-Part I
Artificial Intelligence: A Modern Approach

Artificial Intelligence: A Modern Approach
iTrack: Image-Based Probabilistic Tracking of People

ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 3
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking

IEEE Transactions on Signal Processing

Improving tracking by handling occlusions

ICAPR'05 Proceedings of the Third international conference on Pattern Recognition and Image Analysis - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Condensation is a widely-used tracking algorithm based on particle filters. Although some results have been achieved, it has several unpleasant behaviours. In this paper, we highlight these misbehaviours and propose two improvements. A new weight assignment, which avoids sample impoverishment, is presented. Subsequently, the prediction process is enhanced. The proposal has been successfully tested using synthetic data, which reproduces some of the main difficulties a tracker must deal with.