A Generic Framework for Tracking Using Particle Filter With Dynamic Shape Prior

by: Y Rathi, N Vaswani, A Tannenbaum

Image Processing, IEEE Transactions on, Vol. 16, No. 5. (2007), pp. 1370-1382.

View FullText article

IEEE Explore

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

tyler, willie_gt

Find related articles with these CiteULike tags

filters, kalman, particle, tracking

Abstract

Tracking deforming objects involves estimating the global motion of the object and its local deformations as functions of time. Tracking algorithms using Kalman filters or particle filters (PFs) have been proposed for tracking such objects, but these have limitations due to the lack of dynamic shape information. In this paper, we propose a novel method based on employing a locally linear embedding in order to incorporate dynamic shape information into the particle filtering framework for tracking highly deformable objects in the presence of noise and clutter. The PF also models image statistics such as mean and variance of the given data which can be useful in obtaining proper separation of object and background

@article{citeulike:1598888, abstract = {Tracking deforming objects involves estimating the global motion of the object and its local deformations as functions of time. Tracking algorithms using Kalman filters or particle filters (PFs) have been proposed for tracking such objects, but these have limitations due to the lack of dynamic shape information. In this paper, we propose a novel method based on employing a locally linear embedding in order to incorporate dynamic shape information into the particle filtering framework for tracking highly deformable objects in the presence of noise and clutter. The PF also models image statistics such as mean and variance of the given data which can be useful in obtaining proper separation of object and background}, author = {Rathi, Y. and Vaswani, N. and Tannenbaum, A. }, booktitle = {Image Processing, IEEE Transactions on}, citeulike-article-id = {1598888}, journal = {Image Processing, IEEE Transactions on}, keywords = {kalman, particle, tracking}, number = {5}, pages = {1370--1382}, posted-at = {2007-10-31 16:14:30}, priority = {5}, title = {A Generic Framework for Tracking Using Particle Filter With Dynamic Shape Prior}, url = {http://ieeexplore.ieee.org/xpls/abs\_all.jsp?arnumber=4154802}, volume = {16}, year = {2007} }

RIS record

TY - JOUR ID - citeulike:1598888 L3 - citeulike-article-id:1598888 TI - A Generic Framework for Tracking Using Particle Filter With Dynamic Shape Prior T2 - Image Processing, IEEE Transactions on JF - Image Processing, IEEE Transactions on VL - 16 IS - 5 SP - 1370 EP - 1382 N2 - Tracking deforming objects involves estimating the global motion of the object and its local deformations as functions of time. Tracking algorithms using Kalman filters or particle filters (PFs) have been proposed for tracking such objects, but these have limitations due to the lack of dynamic shape information. In this paper, we propose a novel method based on employing a locally linear embedding in order to incorporate dynamic shape information into the particle filtering framework for tracking highly deformable objects in the presence of noise and clutter. The PF also models image statistics such as mean and variance of the given data which can be useful in obtaining proper separation of object and background KW - kalman KW - particle KW - tracking AU - Rathi, Y AU - Vaswani, N AU - Tannenbaum, A PY - 2007/// UR - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4154802 ER -

RIS BibTeX