Part-based data association for visual tracking

Zhengqiang Jiang, Du Q. Huynh, Jian Zhang, Qiang Wu

Research output: Chapter in Book/Conference paperConference paper

Abstract

We present a method that integrates a part-based sparse appearance model in a Bayesian inference framework for tracking targets in video sequences. We formulate the sparse appearance model as a set of smoothed colour histograms corresponding to the object windows detected by the Deformable Part Model (DPM) detector. The data association of each body part between frames is solved based on the position constraint, appearance coherence, and motion consistency. To deal with missing and noisy observations, the part detection window in the following frame is also predicted using an interacting multiple model (IMM) tracker. We have tested our tracking method on all the video sequences that involve people in upright poses from the TB-50 and TB-100 benchmark videos datasets. Our experimental results show that our tracking method outperforms six state-of-the-art tracking techniques. © 2017 IEEE.
Original languageEnglish
Title of host publication2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages1-8
Number of pages8
ISBN (Electronic)9781538628386
ISBN (Print)9781538628393
DOIs
Publication statusPublished - 21 Dec 2017
Event2017 International Conference on Digital Image Computing: Techniques and Applications - Sydney, Australia
Duration: 29 Nov 20171 Dec 2017

Conference

Conference2017 International Conference on Digital Image Computing: Techniques and Applications
Abbreviated titleDICTA
CountryAustralia
CitySydney
Period29/11/171/12/17

Fingerprint

Target tracking
Color
Detectors

Cite this

Jiang, Z., Huynh, D. Q., Zhang, J., & Wu, Q. (2017). Part-based data association for visual tracking. In 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA) (pp. 1-8). United States: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/DICTA.2017.8227474
Jiang, Zhengqiang ; Huynh, Du Q. ; Zhang, Jian ; Wu, Qiang. / Part-based data association for visual tracking. 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA). United States : IEEE, Institute of Electrical and Electronics Engineers, 2017. pp. 1-8
@inproceedings{f561d3956d154723a856ddd862c589a7,
title = "Part-based data association for visual tracking",
abstract = "We present a method that integrates a part-based sparse appearance model in a Bayesian inference framework for tracking targets in video sequences. We formulate the sparse appearance model as a set of smoothed colour histograms corresponding to the object windows detected by the Deformable Part Model (DPM) detector. The data association of each body part between frames is solved based on the position constraint, appearance coherence, and motion consistency. To deal with missing and noisy observations, the part detection window in the following frame is also predicted using an interacting multiple model (IMM) tracker. We have tested our tracking method on all the video sequences that involve people in upright poses from the TB-50 and TB-100 benchmark videos datasets. Our experimental results show that our tracking method outperforms six state-of-the-art tracking techniques. {\circledC} 2017 IEEE.",
author = "Zhengqiang Jiang and Huynh, {Du Q.} and Jian Zhang and Qiang Wu",
year = "2017",
month = "12",
day = "21",
doi = "10.1109/DICTA.2017.8227474",
language = "English",
isbn = "9781538628393",
pages = "1--8",
booktitle = "2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Jiang, Z, Huynh, DQ, Zhang, J & Wu, Q 2017, Part-based data association for visual tracking. in 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA). IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 1-8, 2017 International Conference on Digital Image Computing: Techniques and Applications, Sydney, Australia, 29/11/17. https://doi.org/10.1109/DICTA.2017.8227474

Part-based data association for visual tracking. / Jiang, Zhengqiang; Huynh, Du Q.; Zhang, Jian; Wu, Qiang.

2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA). United States : IEEE, Institute of Electrical and Electronics Engineers, 2017. p. 1-8.

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Part-based data association for visual tracking

AU - Jiang, Zhengqiang

AU - Huynh, Du Q.

AU - Zhang, Jian

AU - Wu, Qiang

PY - 2017/12/21

Y1 - 2017/12/21

N2 - We present a method that integrates a part-based sparse appearance model in a Bayesian inference framework for tracking targets in video sequences. We formulate the sparse appearance model as a set of smoothed colour histograms corresponding to the object windows detected by the Deformable Part Model (DPM) detector. The data association of each body part between frames is solved based on the position constraint, appearance coherence, and motion consistency. To deal with missing and noisy observations, the part detection window in the following frame is also predicted using an interacting multiple model (IMM) tracker. We have tested our tracking method on all the video sequences that involve people in upright poses from the TB-50 and TB-100 benchmark videos datasets. Our experimental results show that our tracking method outperforms six state-of-the-art tracking techniques. © 2017 IEEE.

AB - We present a method that integrates a part-based sparse appearance model in a Bayesian inference framework for tracking targets in video sequences. We formulate the sparse appearance model as a set of smoothed colour histograms corresponding to the object windows detected by the Deformable Part Model (DPM) detector. The data association of each body part between frames is solved based on the position constraint, appearance coherence, and motion consistency. To deal with missing and noisy observations, the part detection window in the following frame is also predicted using an interacting multiple model (IMM) tracker. We have tested our tracking method on all the video sequences that involve people in upright poses from the TB-50 and TB-100 benchmark videos datasets. Our experimental results show that our tracking method outperforms six state-of-the-art tracking techniques. © 2017 IEEE.

U2 - 10.1109/DICTA.2017.8227474

DO - 10.1109/DICTA.2017.8227474

M3 - Conference paper

SN - 9781538628393

SP - 1

EP - 8

BT - 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA)

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

ER -

Jiang Z, Huynh DQ, Zhang J, Wu Q. Part-based data association for visual tracking. In 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA). United States: IEEE, Institute of Electrical and Electronics Engineers. 2017. p. 1-8 https://doi.org/10.1109/DICTA.2017.8227474