Labeled multi-Bernoulli track-before-detect for multi-target tracking in video

Tharindu Rathnayake; Amirali Khodadadian Gostar; Reza Hoseinnezhad; Alireza Bab-Hadiashar

Labeled multi-Bernoulli track-before-detect for multi-target tracking in video

Tharindu Rathnayake, Amirali Khodadadian Gostar, Reza Hoseinnezhad, Alireza Bab-Hadiashar

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

Abstract

This paper presents a labeled multi-Bernoulli filter for track-before-detect with a special focus on visual tracking of multiple targets in video. We show that labeled multi-Bernoulli distribution is a conjugate prior for an image likelihood function with a specific separable form. Following a previously formulated likelihood function (with the desirable separable form) using background subtraction, we apply our proposed labeled multi-Bernoulli filter. Our simulation results show that the proposed solution can successfully track multiple targets in a public visual tracking dataset. Comparative results show superior tracking performance compared with recent competing methods.

Original language	English
Title of host publication	2015 18th International Conference on Information Fusion, Fusion 2015
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1353-1358
Number of pages	6
ISBN (Electronic)	9780982443866
Publication status	Published - 2015
Externally published	Yes
Event	International Conference on Information Fusion 2015 - Washington, United States of America Duration: 6 Jul 2015 → 9 Jul 2015 Conference number: 18th

Conference

Conference	International Conference on Information Fusion 2015
Abbreviated title	Fusion 2015
Country/Territory	United States of America
City	Washington
Period	6/07/15 → 9/07/15

Cite this

@inproceedings{8ae3f6990f2f430e85ed0e221708bbce,

title = "Labeled multi-Bernoulli track-before-detect for multi-target tracking in video",

abstract = "This paper presents a labeled multi-Bernoulli filter for track-before-detect with a special focus on visual tracking of multiple targets in video. We show that labeled multi-Bernoulli distribution is a conjugate prior for an image likelihood function with a specific separable form. Following a previously formulated likelihood function (with the desirable separable form) using background subtraction, we apply our proposed labeled multi-Bernoulli filter. Our simulation results show that the proposed solution can successfully track multiple targets in a public visual tracking dataset. Comparative results show superior tracking performance compared with recent competing methods.",

author = "Tharindu Rathnayake and Gostar, \{Amirali Khodadadian\} and Reza Hoseinnezhad and Alireza Bab-Hadiashar",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; International Conference on Information Fusion 2015, Fusion 2015 ; Conference date: 06-07-2015 Through 09-07-2015",

year = "2015",

language = "English",

pages = "1353--1358",

booktitle = "2015 18th International Conference on Information Fusion, Fusion 2015",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States of America",

}

Rathnayake, T, Gostar, AK, Hoseinnezhad, R & Bab-Hadiashar, A 2015, Labeled multi-Bernoulli track-before-detect for multi-target tracking in video. in 2015 18th International Conference on Information Fusion, Fusion 2015., 7266714, IEEE, Institute of Electrical and Electronics Engineers, pp. 1353-1358, International Conference on Information Fusion 2015, Washington, District of Columbia, United States of America, 6/07/15.

Labeled multi-Bernoulli track-before-detect for multi-target tracking in video. / Rathnayake, Tharindu; Gostar, Amirali Khodadadian; Hoseinnezhad, Reza et al.
2015 18th International Conference on Information Fusion, Fusion 2015. IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 1353-1358 7266714.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Labeled multi-Bernoulli track-before-detect for multi-target tracking in video

AU - Rathnayake, Tharindu

AU - Gostar, Amirali Khodadadian

AU - Hoseinnezhad, Reza

AU - Bab-Hadiashar, Alireza

N1 - Conference code: 18th

PY - 2015

Y1 - 2015

N2 - This paper presents a labeled multi-Bernoulli filter for track-before-detect with a special focus on visual tracking of multiple targets in video. We show that labeled multi-Bernoulli distribution is a conjugate prior for an image likelihood function with a specific separable form. Following a previously formulated likelihood function (with the desirable separable form) using background subtraction, we apply our proposed labeled multi-Bernoulli filter. Our simulation results show that the proposed solution can successfully track multiple targets in a public visual tracking dataset. Comparative results show superior tracking performance compared with recent competing methods.

AB - This paper presents a labeled multi-Bernoulli filter for track-before-detect with a special focus on visual tracking of multiple targets in video. We show that labeled multi-Bernoulli distribution is a conjugate prior for an image likelihood function with a specific separable form. Following a previously formulated likelihood function (with the desirable separable form) using background subtraction, we apply our proposed labeled multi-Bernoulli filter. Our simulation results show that the proposed solution can successfully track multiple targets in a public visual tracking dataset. Comparative results show superior tracking performance compared with recent competing methods.

UR - https://www.scopus.com/pages/publications/84958201101

M3 - Conference Paper

AN - SCOPUS:84958201101

SP - 1353

EP - 1358

BT - 2015 18th International Conference on Information Fusion, Fusion 2015

PB - IEEE, Institute of Electrical and Electronics Engineers

T2 - International Conference on Information Fusion 2015

Y2 - 6 July 2015 through 9 July 2015

ER -

Labeled multi-Bernoulli track-before-detect for multi-target tracking in video

Abstract

Conference

Other files and links

Cite this