A KLT-based approach for occlusion handling in human tracking

Chenyuan Zhang, Jiu Xu, Axel Beaugendre, Satoshi Goto

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

9 Citations (Scopus)

Abstract

Occlusions significantly affect the result during human tracking. This paper proposes a novel occlusion detection and handling algorithm which is mainly based on the KLT (Kanade-Lucas-Tomasi) method. Instead of using KLT as a tracker, we apply it for occlusion detection to enhance tracking stability. In this paper, a combinational method of particle filter tracking and occlusion detection is proposed. Depending on the detection result, our method makes decisions that whether to update the appearance model and use the occlusion handling strategy. Our occlusion detector associates color information, KLT feature tracker and directions of feature points. Additional, the occlusion handling strategy is based on the information from detection. Moreover, the algorithm also can solve the drift problem. Experimental results on famous datasets prove that our method has better performance and robustness on occlusion detection and handling.

Original languageEnglish
Title of host publication2012 Picture Coding Symposium, PCS 2012, Proceedings
Pages337-340
Number of pages4
DOIs
Publication statusPublished - 2012
Event29th Picture Coding Symposium, PCS 2012 - Krakow
Duration: 2012 May 72012 May 9

Other

Other29th Picture Coding Symposium, PCS 2012
CityKrakow
Period12/5/712/5/9

Fingerprint

Color
Detectors

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Zhang, C., Xu, J., Beaugendre, A., & Goto, S. (2012). A KLT-based approach for occlusion handling in human tracking. In 2012 Picture Coding Symposium, PCS 2012, Proceedings (pp. 337-340). [6213360] https://doi.org/10.1109/PCS.2012.6213360

A KLT-based approach for occlusion handling in human tracking. / Zhang, Chenyuan; Xu, Jiu; Beaugendre, Axel; Goto, Satoshi.

2012 Picture Coding Symposium, PCS 2012, Proceedings. 2012. p. 337-340 6213360.

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

Zhang, C, Xu, J, Beaugendre, A & Goto, S 2012, A KLT-based approach for occlusion handling in human tracking. in 2012 Picture Coding Symposium, PCS 2012, Proceedings., 6213360, pp. 337-340, 29th Picture Coding Symposium, PCS 2012, Krakow, 12/5/7. https://doi.org/10.1109/PCS.2012.6213360
Zhang C, Xu J, Beaugendre A, Goto S. A KLT-based approach for occlusion handling in human tracking. In 2012 Picture Coding Symposium, PCS 2012, Proceedings. 2012. p. 337-340. 6213360 https://doi.org/10.1109/PCS.2012.6213360
Zhang, Chenyuan ; Xu, Jiu ; Beaugendre, Axel ; Goto, Satoshi. / A KLT-based approach for occlusion handling in human tracking. 2012 Picture Coding Symposium, PCS 2012, Proceedings. 2012. pp. 337-340
@inproceedings{4405dcb7f89542d8b47cf7726e98dbb7,
title = "A KLT-based approach for occlusion handling in human tracking",
abstract = "Occlusions significantly affect the result during human tracking. This paper proposes a novel occlusion detection and handling algorithm which is mainly based on the KLT (Kanade-Lucas-Tomasi) method. Instead of using KLT as a tracker, we apply it for occlusion detection to enhance tracking stability. In this paper, a combinational method of particle filter tracking and occlusion detection is proposed. Depending on the detection result, our method makes decisions that whether to update the appearance model and use the occlusion handling strategy. Our occlusion detector associates color information, KLT feature tracker and directions of feature points. Additional, the occlusion handling strategy is based on the information from detection. Moreover, the algorithm also can solve the drift problem. Experimental results on famous datasets prove that our method has better performance and robustness on occlusion detection and handling.",
author = "Chenyuan Zhang and Jiu Xu and Axel Beaugendre and Satoshi Goto",
year = "2012",
doi = "10.1109/PCS.2012.6213360",
language = "English",
isbn = "9781457720482",
pages = "337--340",
booktitle = "2012 Picture Coding Symposium, PCS 2012, Proceedings",

}

TY - GEN

T1 - A KLT-based approach for occlusion handling in human tracking

AU - Zhang, Chenyuan

AU - Xu, Jiu

AU - Beaugendre, Axel

AU - Goto, Satoshi

PY - 2012

Y1 - 2012

N2 - Occlusions significantly affect the result during human tracking. This paper proposes a novel occlusion detection and handling algorithm which is mainly based on the KLT (Kanade-Lucas-Tomasi) method. Instead of using KLT as a tracker, we apply it for occlusion detection to enhance tracking stability. In this paper, a combinational method of particle filter tracking and occlusion detection is proposed. Depending on the detection result, our method makes decisions that whether to update the appearance model and use the occlusion handling strategy. Our occlusion detector associates color information, KLT feature tracker and directions of feature points. Additional, the occlusion handling strategy is based on the information from detection. Moreover, the algorithm also can solve the drift problem. Experimental results on famous datasets prove that our method has better performance and robustness on occlusion detection and handling.

AB - Occlusions significantly affect the result during human tracking. This paper proposes a novel occlusion detection and handling algorithm which is mainly based on the KLT (Kanade-Lucas-Tomasi) method. Instead of using KLT as a tracker, we apply it for occlusion detection to enhance tracking stability. In this paper, a combinational method of particle filter tracking and occlusion detection is proposed. Depending on the detection result, our method makes decisions that whether to update the appearance model and use the occlusion handling strategy. Our occlusion detector associates color information, KLT feature tracker and directions of feature points. Additional, the occlusion handling strategy is based on the information from detection. Moreover, the algorithm also can solve the drift problem. Experimental results on famous datasets prove that our method has better performance and robustness on occlusion detection and handling.

UR - http://www.scopus.com/inward/record.url?scp=84864067892&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864067892&partnerID=8YFLogxK

U2 - 10.1109/PCS.2012.6213360

DO - 10.1109/PCS.2012.6213360

M3 - Conference contribution

AN - SCOPUS:84864067892

SN - 9781457720482

SP - 337

EP - 340

BT - 2012 Picture Coding Symposium, PCS 2012, Proceedings

ER -