3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis

Yang Liu, Shuyi Huang, Xina Cheng, Takeshi Ikenaga

研究成果: Conference contribution

抄録

Volleyball video analysis is important for developing applications such as player evaluation system or tactic analysis system. Among its different topics, player action recognition serves as an elementary building brick for understanding player’s behavior. Most conventional player action recognition methods have limits in real volleyball game due to the occlusion and intra-class variation problems. This paper proposes a 3D global trajectory and multi-view local motion combined volleyball player action recognition method. 3D global trajectory extracts global motion feature through 3D trajectories, which hides the unstable and incomplete 2D motion feature caused by the above problems. Multi-view local motion gets detailed local motion feature of arms and legs in multiple viewpoints and removes clutter features caused by occlusion problem. Through the combination, global 3D feature and local motion feature mutually promote each other and the actions are recognized well. Experiments are conducted on game videos from the Semifinal and Final Game of 2014 Japan Inter High School Games of Men’s Volleyball in Tokyo Metropolitan Gymnasium. The experiments show the combing result accuracy achieves 98.39%, 95.50%, 96.86%, 96.98% for spike, block, receive, toss respectively and improve 11.33% averagely than the sing-view local motion based result.

元の言語English
ホスト出版物のタイトルAdvances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings
編集者Chong-Wah Ngo, Toshihiko Yamasaki, Richang Hong, Meng Wang, Wen-Huang Cheng
出版者Springer-Verlag
ページ134-144
ページ数11
ISBN(印刷物)9783030007638
DOI
出版物ステータスPublished - 2018 1 1
イベント19th Pacific-Rim Conference on Multimedia, PCM 2018 - Hefei, China
継続期間: 2018 9 212018 9 22

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
11166 LNCS
ISSN(印刷物)0302-9743
ISSN(電子版)1611-3349

Other

Other19th Pacific-Rim Conference on Multimedia, PCM 2018
China
Hefei
期間18/9/2118/9/22

Fingerprint

Action Recognition
Trajectories
Trajectory
Motion
Brick buildings
Game
Occlusion
Experiments
Video Analysis
Video Games
Clutter
Systems Analysis
Spike
Japan
Experiment
Unstable
Evaluation

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

これを引用

Liu, Y., Huang, S., Cheng, X., & Ikenaga, T. (2018). 3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis. : C-W. Ngo, T. Yamasaki, R. Hong, M. Wang, & W-H. Cheng (版), Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings (pp. 134-144). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 巻数 11166 LNCS). Springer-Verlag. https://doi.org/10.1007/978-3-030-00764-5_13

3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis. / Liu, Yang; Huang, Shuyi; Cheng, Xina; Ikenaga, Takeshi.

Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings. 版 / Chong-Wah Ngo; Toshihiko Yamasaki; Richang Hong; Meng Wang; Wen-Huang Cheng. Springer-Verlag, 2018. p. 134-144 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 巻 11166 LNCS).

研究成果: Conference contribution

Liu, Y, Huang, S, Cheng, X & Ikenaga, T 2018, 3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis. : C-W Ngo, T Yamasaki, R Hong, M Wang & W-H Cheng (版), Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 巻. 11166 LNCS, Springer-Verlag, pp. 134-144, 19th Pacific-Rim Conference on Multimedia, PCM 2018, Hefei, China, 18/9/21. https://doi.org/10.1007/978-3-030-00764-5_13
Liu Y, Huang S, Cheng X, Ikenaga T. 3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis. : Ngo C-W, Yamasaki T, Hong R, Wang M, Cheng W-H, 編集者, Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings. Springer-Verlag. 2018. p. 134-144. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-00764-5_13
Liu, Yang ; Huang, Shuyi ; Cheng, Xina ; Ikenaga, Takeshi. / 3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis. Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings. 編集者 / Chong-Wah Ngo ; Toshihiko Yamasaki ; Richang Hong ; Meng Wang ; Wen-Huang Cheng. Springer-Verlag, 2018. pp. 134-144 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{c861f5f5ab7543049955cf1300a35930,
title = "3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis",
abstract = "Volleyball video analysis is important for developing applications such as player evaluation system or tactic analysis system. Among its different topics, player action recognition serves as an elementary building brick for understanding player’s behavior. Most conventional player action recognition methods have limits in real volleyball game due to the occlusion and intra-class variation problems. This paper proposes a 3D global trajectory and multi-view local motion combined volleyball player action recognition method. 3D global trajectory extracts global motion feature through 3D trajectories, which hides the unstable and incomplete 2D motion feature caused by the above problems. Multi-view local motion gets detailed local motion feature of arms and legs in multiple viewpoints and removes clutter features caused by occlusion problem. Through the combination, global 3D feature and local motion feature mutually promote each other and the actions are recognized well. Experiments are conducted on game videos from the Semifinal and Final Game of 2014 Japan Inter High School Games of Men’s Volleyball in Tokyo Metropolitan Gymnasium. The experiments show the combing result accuracy achieves 98.39{\%}, 95.50{\%}, 96.86{\%}, 96.98{\%} for spike, block, receive, toss respectively and improve 11.33{\%} averagely than the sing-view local motion based result.",
keywords = "Intra-class variation, Occlusion, Player action recognition, Volleyball analysis",
author = "Yang Liu and Shuyi Huang and Xina Cheng and Takeshi Ikenaga",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-00764-5_13",
language = "English",
isbn = "9783030007638",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "134--144",
editor = "Chong-Wah Ngo and Toshihiko Yamasaki and Richang Hong and Meng Wang and Wen-Huang Cheng",
booktitle = "Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings",

}

TY - GEN

T1 - 3D global trajectory and multi-view local motion combined player action recognition in volleyball analysis

AU - Liu, Yang

AU - Huang, Shuyi

AU - Cheng, Xina

AU - Ikenaga, Takeshi

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Volleyball video analysis is important for developing applications such as player evaluation system or tactic analysis system. Among its different topics, player action recognition serves as an elementary building brick for understanding player’s behavior. Most conventional player action recognition methods have limits in real volleyball game due to the occlusion and intra-class variation problems. This paper proposes a 3D global trajectory and multi-view local motion combined volleyball player action recognition method. 3D global trajectory extracts global motion feature through 3D trajectories, which hides the unstable and incomplete 2D motion feature caused by the above problems. Multi-view local motion gets detailed local motion feature of arms and legs in multiple viewpoints and removes clutter features caused by occlusion problem. Through the combination, global 3D feature and local motion feature mutually promote each other and the actions are recognized well. Experiments are conducted on game videos from the Semifinal and Final Game of 2014 Japan Inter High School Games of Men’s Volleyball in Tokyo Metropolitan Gymnasium. The experiments show the combing result accuracy achieves 98.39%, 95.50%, 96.86%, 96.98% for spike, block, receive, toss respectively and improve 11.33% averagely than the sing-view local motion based result.

AB - Volleyball video analysis is important for developing applications such as player evaluation system or tactic analysis system. Among its different topics, player action recognition serves as an elementary building brick for understanding player’s behavior. Most conventional player action recognition methods have limits in real volleyball game due to the occlusion and intra-class variation problems. This paper proposes a 3D global trajectory and multi-view local motion combined volleyball player action recognition method. 3D global trajectory extracts global motion feature through 3D trajectories, which hides the unstable and incomplete 2D motion feature caused by the above problems. Multi-view local motion gets detailed local motion feature of arms and legs in multiple viewpoints and removes clutter features caused by occlusion problem. Through the combination, global 3D feature and local motion feature mutually promote each other and the actions are recognized well. Experiments are conducted on game videos from the Semifinal and Final Game of 2014 Japan Inter High School Games of Men’s Volleyball in Tokyo Metropolitan Gymnasium. The experiments show the combing result accuracy achieves 98.39%, 95.50%, 96.86%, 96.98% for spike, block, receive, toss respectively and improve 11.33% averagely than the sing-view local motion based result.

KW - Intra-class variation

KW - Occlusion

KW - Player action recognition

KW - Volleyball analysis

UR - http://www.scopus.com/inward/record.url?scp=85054492166&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054492166&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00764-5_13

DO - 10.1007/978-3-030-00764-5_13

M3 - Conference contribution

AN - SCOPUS:85054492166

SN - 9783030007638

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 134

EP - 144

BT - Advances in Multimedia Information Processing – PCM 2018 - 19th Pacific-Rim Conference on Multimedia, 2018, Proceedings

A2 - Ngo, Chong-Wah

A2 - Yamasaki, Toshihiko

A2 - Hong, Richang

A2 - Wang, Meng

A2 - Cheng, Wen-Huang

PB - Springer-Verlag

ER -