Robust multi-object tracking via cross-domain contextual information for sports video analysis

Tianzhu Zhang, Bernard Ghanem, Narendra Ahuja

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Multiple player tracking is one of the main building blocks needed in a sports video analysis system. In an uncalibrated camera setting, robust mutli-object tracking can be very difficult due to a number of reasons including the presence of noise, occlusion, fast camera motion, low-resolution image capture, varying viewpoints and illumination changes. To address the problem of multi-object tracking in sports videos, we go beyond the video frame domain and make use of information in a homography transform domain that is denoted the homography field domain. We propose a novel particle filter based tracking algorithm that uses both object appearance information (e.g. color and shape) in the image domain and cross-domain contextual information in the field domain to improve object tracking. In the field domain, the effect of fast camera motion is significantly alleviated since the underlying homography transform from each frame to the field domain can be accurately estimated. We use contextual trajectory information (intra-trajectory and inter-trajectory context) to further improve the prediction of object states within an particle filter framework. Here, intra-trajectory contextual information is based on history tracking results in the field domain, while inter-trajectory contextual information is extracted from a compiled trajectory dataset based on tracks computed from videos depicting the same sport. Experimental results on real world sports data show that our system is able to effectively and robustly track a variable number of targets regardless of background clutter, camera motion and frequent mutual occlusion between targets.

Original languageEnglish (US)
Title of host publication2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings
Pages985-988
Number of pages4
DOIs
StatePublished - Oct 23 2012
Event2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Kyoto, Japan
Duration: Mar 25 2012Mar 30 2012

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)1520-6149

Other

Other2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012
CountryJapan
CityKyoto
Period3/25/123/30/12

Fingerprint

Sports
Trajectories
Cameras
Image resolution
Lighting
Color

Keywords

  • Contextual Information
  • Cross-Domain
  • Particle Filter
  • Tracking

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Zhang, T., Ghanem, B., & Ahuja, N. (2012). Robust multi-object tracking via cross-domain contextual information for sports video analysis. In 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings (pp. 985-988). [6288050] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2012.6288050

Robust multi-object tracking via cross-domain contextual information for sports video analysis. / Zhang, Tianzhu; Ghanem, Bernard; Ahuja, Narendra.

2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings. 2012. p. 985-988 6288050 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, T, Ghanem, B & Ahuja, N 2012, Robust multi-object tracking via cross-domain contextual information for sports video analysis. in 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings., 6288050, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, pp. 985-988, 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012, Kyoto, Japan, 3/25/12. https://doi.org/10.1109/ICASSP.2012.6288050
Zhang T, Ghanem B, Ahuja N. Robust multi-object tracking via cross-domain contextual information for sports video analysis. In 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings. 2012. p. 985-988. 6288050. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2012.6288050
Zhang, Tianzhu ; Ghanem, Bernard ; Ahuja, Narendra. / Robust multi-object tracking via cross-domain contextual information for sports video analysis. 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings. 2012. pp. 985-988 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{9b9a02255a8948098b5eb0122d3e36dc,
title = "Robust multi-object tracking via cross-domain contextual information for sports video analysis",
abstract = "Multiple player tracking is one of the main building blocks needed in a sports video analysis system. In an uncalibrated camera setting, robust mutli-object tracking can be very difficult due to a number of reasons including the presence of noise, occlusion, fast camera motion, low-resolution image capture, varying viewpoints and illumination changes. To address the problem of multi-object tracking in sports videos, we go beyond the video frame domain and make use of information in a homography transform domain that is denoted the homography field domain. We propose a novel particle filter based tracking algorithm that uses both object appearance information (e.g. color and shape) in the image domain and cross-domain contextual information in the field domain to improve object tracking. In the field domain, the effect of fast camera motion is significantly alleviated since the underlying homography transform from each frame to the field domain can be accurately estimated. We use contextual trajectory information (intra-trajectory and inter-trajectory context) to further improve the prediction of object states within an particle filter framework. Here, intra-trajectory contextual information is based on history tracking results in the field domain, while inter-trajectory contextual information is extracted from a compiled trajectory dataset based on tracks computed from videos depicting the same sport. Experimental results on real world sports data show that our system is able to effectively and robustly track a variable number of targets regardless of background clutter, camera motion and frequent mutual occlusion between targets.",
keywords = "Contextual Information, Cross-Domain, Particle Filter, Tracking",
author = "Tianzhu Zhang and Bernard Ghanem and Narendra Ahuja",
year = "2012",
month = "10",
day = "23",
doi = "10.1109/ICASSP.2012.6288050",
language = "English (US)",
isbn = "9781467300469",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
pages = "985--988",
booktitle = "2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings",

}

TY - GEN

T1 - Robust multi-object tracking via cross-domain contextual information for sports video analysis

AU - Zhang, Tianzhu

AU - Ghanem, Bernard

AU - Ahuja, Narendra

PY - 2012/10/23

Y1 - 2012/10/23

N2 - Multiple player tracking is one of the main building blocks needed in a sports video analysis system. In an uncalibrated camera setting, robust mutli-object tracking can be very difficult due to a number of reasons including the presence of noise, occlusion, fast camera motion, low-resolution image capture, varying viewpoints and illumination changes. To address the problem of multi-object tracking in sports videos, we go beyond the video frame domain and make use of information in a homography transform domain that is denoted the homography field domain. We propose a novel particle filter based tracking algorithm that uses both object appearance information (e.g. color and shape) in the image domain and cross-domain contextual information in the field domain to improve object tracking. In the field domain, the effect of fast camera motion is significantly alleviated since the underlying homography transform from each frame to the field domain can be accurately estimated. We use contextual trajectory information (intra-trajectory and inter-trajectory context) to further improve the prediction of object states within an particle filter framework. Here, intra-trajectory contextual information is based on history tracking results in the field domain, while inter-trajectory contextual information is extracted from a compiled trajectory dataset based on tracks computed from videos depicting the same sport. Experimental results on real world sports data show that our system is able to effectively and robustly track a variable number of targets regardless of background clutter, camera motion and frequent mutual occlusion between targets.

AB - Multiple player tracking is one of the main building blocks needed in a sports video analysis system. In an uncalibrated camera setting, robust mutli-object tracking can be very difficult due to a number of reasons including the presence of noise, occlusion, fast camera motion, low-resolution image capture, varying viewpoints and illumination changes. To address the problem of multi-object tracking in sports videos, we go beyond the video frame domain and make use of information in a homography transform domain that is denoted the homography field domain. We propose a novel particle filter based tracking algorithm that uses both object appearance information (e.g. color and shape) in the image domain and cross-domain contextual information in the field domain to improve object tracking. In the field domain, the effect of fast camera motion is significantly alleviated since the underlying homography transform from each frame to the field domain can be accurately estimated. We use contextual trajectory information (intra-trajectory and inter-trajectory context) to further improve the prediction of object states within an particle filter framework. Here, intra-trajectory contextual information is based on history tracking results in the field domain, while inter-trajectory contextual information is extracted from a compiled trajectory dataset based on tracks computed from videos depicting the same sport. Experimental results on real world sports data show that our system is able to effectively and robustly track a variable number of targets regardless of background clutter, camera motion and frequent mutual occlusion between targets.

KW - Contextual Information

KW - Cross-Domain

KW - Particle Filter

KW - Tracking

UR - http://www.scopus.com/inward/record.url?scp=84867597162&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867597162&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2012.6288050

DO - 10.1109/ICASSP.2012.6288050

M3 - Conference contribution

AN - SCOPUS:84867597162

SN - 9781467300469

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 985

EP - 988

BT - 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Proceedings

ER -