On sequential Monte Carlo sampling methods for Bayesian filtering
Statistics and Computing
Color-Based Probabilistic Tracking
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Multimodal Shape Tracking with Point Distribution Models
Proceedings of the 24th DAGM Symposium on Pattern Recognition
Maintaining Multi-Modality through Mixture Tracking
ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
ACM Computing Surveys (CSUR)
Variable-mass particle filter for road-constrained vehicle tracking
EURASIP Journal on Advances in Signal Processing
A tutorial on particle filters for online nonlinear/non-GaussianBayesian tracking
IEEE Transactions on Signal Processing
Hi-index | 0.00 |
Object tracking in computer vision is an attractive research field due to its widespread application area and challenges. In the recent years, Particle filter is known as a prominent solution for the state estimation problems in point tracking and successfully applied in a wide range of applications. But one of its limitations is the weakness at constantly maintaining the multi-modal target distribution that may arise due to occlusion, clutter or the presence of multiple objects. Lately, that weak point has been overcome in a multi-modal Particle filter (MPF). This paper aims to build some most basic functions of a motorcycle surveillance system using MPF and basing on the color observation model. Accompanied with a simple but effective detecting strategy, the application has the processing rate equivalent to a real time tracking system and high performance.