Particle filter has attracted increasing attention from researchers of object tracking due to its promising property of handling nonlinear and non-Gaussian systems. In this paper, we mainly explore the problem of precisely estimating observation likelihoods of particles in the joint feature-spatial space. For this purpose, a mixture Gaussian kernel function based similarity is presented to evaluate the discrepancy between the target region and the particle region. Such a similarity can be interpreted as the expectation of the spatial weighted feature distribution over the target region. To adapt outburst of object motion, we also present a method to appropriately adjust state transition model by utilizing the priors of motion speed and object size. In comparison with the standard particle filter tracker, our tracking algorithm shows the better performance on challenging video sequences.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Liang SHA, Guijin WANG, Anbang YAO, Xinggang LIN, "Measuring Particles in Joint Feature-Spatial Space" in IEICE TRANSACTIONS on Fundamentals,
vol. E92-A, no. 7, pp. 1737-1742, July 2009, doi: 10.1587/transfun.E92.A.1737.
Abstract: Particle filter has attracted increasing attention from researchers of object tracking due to its promising property of handling nonlinear and non-Gaussian systems. In this paper, we mainly explore the problem of precisely estimating observation likelihoods of particles in the joint feature-spatial space. For this purpose, a mixture Gaussian kernel function based similarity is presented to evaluate the discrepancy between the target region and the particle region. Such a similarity can be interpreted as the expectation of the spatial weighted feature distribution over the target region. To adapt outburst of object motion, we also present a method to appropriately adjust state transition model by utilizing the priors of motion speed and object size. In comparison with the standard particle filter tracker, our tracking algorithm shows the better performance on challenging video sequences.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E92.A.1737/_p
Copy
@ARTICLE{e92-a_7_1737,
author={Liang SHA, Guijin WANG, Anbang YAO, Xinggang LIN, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Measuring Particles in Joint Feature-Spatial Space},
year={2009},
volume={E92-A},
number={7},
pages={1737-1742},
abstract={Particle filter has attracted increasing attention from researchers of object tracking due to its promising property of handling nonlinear and non-Gaussian systems. In this paper, we mainly explore the problem of precisely estimating observation likelihoods of particles in the joint feature-spatial space. For this purpose, a mixture Gaussian kernel function based similarity is presented to evaluate the discrepancy between the target region and the particle region. Such a similarity can be interpreted as the expectation of the spatial weighted feature distribution over the target region. To adapt outburst of object motion, we also present a method to appropriately adjust state transition model by utilizing the priors of motion speed and object size. In comparison with the standard particle filter tracker, our tracking algorithm shows the better performance on challenging video sequences.},
keywords={},
doi={10.1587/transfun.E92.A.1737},
ISSN={1745-1337},
month={July},}
Copy
TY - JOUR
TI - Measuring Particles in Joint Feature-Spatial Space
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1737
EP - 1742
AU - Liang SHA
AU - Guijin WANG
AU - Anbang YAO
AU - Xinggang LIN
PY - 2009
DO - 10.1587/transfun.E92.A.1737
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E92-A
IS - 7
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - July 2009
AB - Particle filter has attracted increasing attention from researchers of object tracking due to its promising property of handling nonlinear and non-Gaussian systems. In this paper, we mainly explore the problem of precisely estimating observation likelihoods of particles in the joint feature-spatial space. For this purpose, a mixture Gaussian kernel function based similarity is presented to evaluate the discrepancy between the target region and the particle region. Such a similarity can be interpreted as the expectation of the spatial weighted feature distribution over the target region. To adapt outburst of object motion, we also present a method to appropriately adjust state transition model by utilizing the priors of motion speed and object size. In comparison with the standard particle filter tracker, our tracking algorithm shows the better performance on challenging video sequences.
ER -