Player detection using one-class SVM

Xuefeng Bai; Tiejun Zhang; Chuanjun Wang; Qiong Li; Xiamu Niu

doi:10.1117/12.956462

Player detection using one-class SVM

Xuefeng Bai^*
, Tiejun Zhang
, Chuanjun Wang
, Qiong Li
, Xiamu Niu

^*Corresponding author for this work

Harbin Institute of Technology Shenzhen
School of Computer Science and Technology, Harbin Institute of Technology

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

Abstract

In this paper, a novel player detection method via One-Class SVM(OCSVM) is proposed, inspired by both the player detection problem and the property of the OCSVM. In this detection method, candidate regions are got by local entropy and local range analysis firstly. Then a set of training samples is obtained by several predefined rules on shape and area. These samples are used to train two OCSVM models. One model uses color feature, and the other uses gradient feature. Finally, we locate the regions of player by fusing the detection result of the two models. Extensive experiments demonstrate effectiveness and efficiency of the proposed method.

Original language	English
Title of host publication	Fourth International Conference on Digital Image Processing, ICDIP 2012
DOIs	https://doi.org/10.1117/12.956462
State	Published - 2012
Externally published	Yes
Event	4th International Conference on Digital Image Processing, ICDIP 2012 - Kuala Lumpur, Malaysia Duration: 7 Apr 2012 → 8 Apr 2012

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	8334
ISSN (Print)	0277-786X

Conference

Conference	4th International Conference on Digital Image Processing, ICDIP 2012
Country/Territory	Malaysia
City	Kuala Lumpur
Period	7/04/12 → 8/04/12

Keywords

one-class support vector machine
player detection
sports video

Access to Document

10.1117/12.956462

Cite this

@inproceedings{7699e5eec39e4fdcb0b74ba7e5ae4ac6,

title = "Player detection using one-class SVM",

abstract = "In this paper, a novel player detection method via One-Class SVM(OCSVM) is proposed, inspired by both the player detection problem and the property of the OCSVM. In this detection method, candidate regions are got by local entropy and local range analysis firstly. Then a set of training samples is obtained by several predefined rules on shape and area. These samples are used to train two OCSVM models. One model uses color feature, and the other uses gradient feature. Finally, we locate the regions of player by fusing the detection result of the two models. Extensive experiments demonstrate effectiveness and efficiency of the proposed method.",

keywords = "one-class support vector machine, player detection, sports video",

author = "Xuefeng Bai and Tiejun Zhang and Chuanjun Wang and Qiong Li and Xiamu Niu",

year = "2012",

doi = "10.1117/12.956462",

language = "英语",

isbn = "9780819489913",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

booktitle = "Fourth International Conference on Digital Image Processing, ICDIP 2012",

note = "4th International Conference on Digital Image Processing, ICDIP 2012 ; Conference date: 07-04-2012 Through 08-04-2012",

}

Bai, X, Zhang, T, Wang, C, Li, Q & Niu, X 2012, Player detection using one-class SVM. in Fourth International Conference on Digital Image Processing, ICDIP 2012., 83342X, Proceedings of SPIE - The International Society for Optical Engineering, vol. 8334, 4th International Conference on Digital Image Processing, ICDIP 2012, Kuala Lumpur, Malaysia, 7/04/12. https://doi.org/10.1117/12.956462

TY - GEN

T1 - Player detection using one-class SVM

AU - Bai, Xuefeng

AU - Zhang, Tiejun

AU - Wang, Chuanjun

AU - Li, Qiong

AU - Niu, Xiamu

PY - 2012

Y1 - 2012

N2 - In this paper, a novel player detection method via One-Class SVM(OCSVM) is proposed, inspired by both the player detection problem and the property of the OCSVM. In this detection method, candidate regions are got by local entropy and local range analysis firstly. Then a set of training samples is obtained by several predefined rules on shape and area. These samples are used to train two OCSVM models. One model uses color feature, and the other uses gradient feature. Finally, we locate the regions of player by fusing the detection result of the two models. Extensive experiments demonstrate effectiveness and efficiency of the proposed method.

AB - In this paper, a novel player detection method via One-Class SVM(OCSVM) is proposed, inspired by both the player detection problem and the property of the OCSVM. In this detection method, candidate regions are got by local entropy and local range analysis firstly. Then a set of training samples is obtained by several predefined rules on shape and area. These samples are used to train two OCSVM models. One model uses color feature, and the other uses gradient feature. Finally, we locate the regions of player by fusing the detection result of the two models. Extensive experiments demonstrate effectiveness and efficiency of the proposed method.

KW - one-class support vector machine

KW - player detection

KW - sports video

UR - https://www.scopus.com/pages/publications/84874726608

U2 - 10.1117/12.956462

DO - 10.1117/12.956462

M3 - 会议稿件

AN - SCOPUS:84874726608

SN - 9780819489913

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Fourth International Conference on Digital Image Processing, ICDIP 2012

T2 - 4th International Conference on Digital Image Processing, ICDIP 2012

Y2 - 7 April 2012 through 8 April 2012

ER -

Player detection using one-class SVM

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this