Spatio-temporal Gaussian process models for extended and group object tracking with irregular shapes

Tutkimustuotos: Lehtiartikkelivertaisarvioitu

Standard

Spatio-temporal Gaussian process models for extended and group object tracking with irregular shapes. / Aftab, Waqas; Hostettler, Roland; De Freitas, Allan; Arvaneh, Mahnaz; Mihaylova, Lyudmila.

julkaisussa: IEEE Transactions on Vehicular Technology, Vuosikerta 68, Nro 3, 8601344, 01.03.2019, s. 2137-2151.

Tutkimustuotos: Lehtiartikkelivertaisarvioitu

Harvard

Aftab, W, Hostettler, R, De Freitas, A, Arvaneh, M & Mihaylova, L 2019, 'Spatio-temporal Gaussian process models for extended and group object tracking with irregular shapes' IEEE Transactions on Vehicular Technology, Vuosikerta. 68, Nro 3, 8601344, Sivut 2137-2151. https://doi.org/10.1109/TVT.2019.2891006

APA

Vancouver

Author

Aftab, Waqas ; Hostettler, Roland ; De Freitas, Allan ; Arvaneh, Mahnaz ; Mihaylova, Lyudmila. / Spatio-temporal Gaussian process models for extended and group object tracking with irregular shapes. Julkaisussa: IEEE Transactions on Vehicular Technology. 2019 ; Vuosikerta 68, Nro 3. Sivut 2137-2151.

Bibtex - Lataa

@article{5388e76f44b2498f8e1f91241231a5c7,
title = "Spatio-temporal Gaussian process models for extended and group object tracking with irregular shapes",
abstract = "Extended object tracking has become an integral part of many autonomous systems during the last two decades. For the first time, this paper presents a generic spatio-temporal Gaussian process (STGP) for tracking an irregular and non-rigid extended object. The complex shape is represented by key points and their parameters are estimated both in space and time. This is achieved by a factorization of the power spectral density function of the STGP covariance function. A new form of the temporal covariance kernel is derived with the theoretical expression of the filter likelihood function. Solutions to both the filtering and the smoothing problems are presented. A thorough evaluation of the performance in a simulated environment shows that the proposed STGP approach outperforms the state-of-the-art GP extended Kalman filter approach [N. Wahlstr{\"o}m and E. {\"O}zkan, 'Extended target tracking using Gaussian processes, IEEE Transactions on Signal Processing,' vol. 63, no. 16, pp. 4165-4178, Aug. 2015] with up to 90{\%} improvement in the accuracy in position, 95{\%} in velocity and 7{\%} in the shape, while tracking a simulated asymmetric non-rigid object. The tracking performance improvement for a non-rigid irregular real object is up to 43{\%} in position, 68{\%} in velocity, 10{\%} in the recall, and 115{\%} in the precision measures.",
keywords = "Extended Object Tracking, Spatio-temporal Gaussian Process, Rauch-Tung-Striebel smoothing",
author = "Waqas Aftab and Roland Hostettler and {De Freitas}, Allan and Mahnaz Arvaneh and Lyudmila Mihaylova",
note = "| openaire: EC/H2020/688082/EU//SETA",
year = "2019",
month = "3",
day = "1",
doi = "10.1109/TVT.2019.2891006",
language = "English",
volume = "68",
pages = "2137--2151",
journal = "IEEE Transactions on Vehicular Technology",
issn = "0018-9545",
number = "3",

}

RIS - Lataa

TY - JOUR

T1 - Spatio-temporal Gaussian process models for extended and group object tracking with irregular shapes

AU - Aftab, Waqas

AU - Hostettler, Roland

AU - De Freitas, Allan

AU - Arvaneh, Mahnaz

AU - Mihaylova, Lyudmila

N1 - | openaire: EC/H2020/688082/EU//SETA

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Extended object tracking has become an integral part of many autonomous systems during the last two decades. For the first time, this paper presents a generic spatio-temporal Gaussian process (STGP) for tracking an irregular and non-rigid extended object. The complex shape is represented by key points and their parameters are estimated both in space and time. This is achieved by a factorization of the power spectral density function of the STGP covariance function. A new form of the temporal covariance kernel is derived with the theoretical expression of the filter likelihood function. Solutions to both the filtering and the smoothing problems are presented. A thorough evaluation of the performance in a simulated environment shows that the proposed STGP approach outperforms the state-of-the-art GP extended Kalman filter approach [N. Wahlström and E. Özkan, 'Extended target tracking using Gaussian processes, IEEE Transactions on Signal Processing,' vol. 63, no. 16, pp. 4165-4178, Aug. 2015] with up to 90% improvement in the accuracy in position, 95% in velocity and 7% in the shape, while tracking a simulated asymmetric non-rigid object. The tracking performance improvement for a non-rigid irregular real object is up to 43% in position, 68% in velocity, 10% in the recall, and 115% in the precision measures.

AB - Extended object tracking has become an integral part of many autonomous systems during the last two decades. For the first time, this paper presents a generic spatio-temporal Gaussian process (STGP) for tracking an irregular and non-rigid extended object. The complex shape is represented by key points and their parameters are estimated both in space and time. This is achieved by a factorization of the power spectral density function of the STGP covariance function. A new form of the temporal covariance kernel is derived with the theoretical expression of the filter likelihood function. Solutions to both the filtering and the smoothing problems are presented. A thorough evaluation of the performance in a simulated environment shows that the proposed STGP approach outperforms the state-of-the-art GP extended Kalman filter approach [N. Wahlström and E. Özkan, 'Extended target tracking using Gaussian processes, IEEE Transactions on Signal Processing,' vol. 63, no. 16, pp. 4165-4178, Aug. 2015] with up to 90% improvement in the accuracy in position, 95% in velocity and 7% in the shape, while tracking a simulated asymmetric non-rigid object. The tracking performance improvement for a non-rigid irregular real object is up to 43% in position, 68% in velocity, 10% in the recall, and 115% in the precision measures.

KW - Extended Object Tracking

KW - Spatio-temporal Gaussian Process

KW - Rauch-Tung-Striebel smoothing

UR - http://www.scopus.com/inward/record.url?scp=85063299199&partnerID=8YFLogxK

U2 - 10.1109/TVT.2019.2891006

DO - 10.1109/TVT.2019.2891006

M3 - Article

VL - 68

SP - 2137

EP - 2151

JO - IEEE Transactions on Vehicular Technology

JF - IEEE Transactions on Vehicular Technology

SN - 0018-9545

IS - 3

M1 - 8601344

ER -

ID: 31067102