Action detection of volleyball using features based on clustering of body trajectories

Eijiro Kubota, Takahiro Suzuki, Masaaki Honda, Takeshi Ikenaga

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

For creating new tactics of sports like volleyball, the analysis of player motion in real games becomes more and more important. However, since motion data needed for the analysis is captured by human observation currently, an automatic capturing system from video camera is highly expected to gather many useful data easily. This paper proposes an action detection algorithm of volleyball players using motion features based on clustering and aggregation of body trajectories. Since the body trajectories of arms and legs are similar, the clustering utilizes shape, location and density of their trajectories. Furthermore, the clustered feature values are aggregated by means of their mean and variance. Experimental results by using the motion detection system based on the proposed algorithm show that it averagely attains 0.9539 AUC of the ROC curve for the detection of four basic motions (block, receive, spike and toss) from the volleyball game video captured by high-definition cameras. This is 0.014775 higher than conventional methods.

Original languageEnglish
Pages (from-to)373-381
Number of pages9
JournalJournal of the Institute of Image Electronics Engineers of Japan
Volume45
Issue number3
DOIs
Publication statusPublished - 2016 Jan 1

Fingerprint

Trajectories
Video cameras
Sports
Agglomeration
Cameras

Keywords

  • Clustering
  • Dense trajectories
  • Feature value aggregation
  • Video detection
  • Volleyball

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Electrical and Electronic Engineering

Cite this

Action detection of volleyball using features based on clustering of body trajectories. / Kubota, Eijiro; Suzuki, Takahiro; Honda, Masaaki; Ikenaga, Takeshi.

In: Journal of the Institute of Image Electronics Engineers of Japan, Vol. 45, No. 3, 01.01.2016, p. 373-381.

Research output: Contribution to journalArticle

@article{04d55dd863344764b5bac381f7e14536,
title = "Action detection of volleyball using features based on clustering of body trajectories",
abstract = "For creating new tactics of sports like volleyball, the analysis of player motion in real games becomes more and more important. However, since motion data needed for the analysis is captured by human observation currently, an automatic capturing system from video camera is highly expected to gather many useful data easily. This paper proposes an action detection algorithm of volleyball players using motion features based on clustering and aggregation of body trajectories. Since the body trajectories of arms and legs are similar, the clustering utilizes shape, location and density of their trajectories. Furthermore, the clustered feature values are aggregated by means of their mean and variance. Experimental results by using the motion detection system based on the proposed algorithm show that it averagely attains 0.9539 AUC of the ROC curve for the detection of four basic motions (block, receive, spike and toss) from the volleyball game video captured by high-definition cameras. This is 0.014775 higher than conventional methods.",
keywords = "Clustering, Dense trajectories, Feature value aggregation, Video detection, Volleyball",
author = "Eijiro Kubota and Takahiro Suzuki and Masaaki Honda and Takeshi Ikenaga",
year = "2016",
month = "1",
day = "1",
doi = "10.11371/iieej.45.373",
language = "English",
volume = "45",
pages = "373--381",
journal = "Journal of the Institute of Image Electronics Engineers of Japan",
issn = "0285-9831",
publisher = "Institute of Image Electronics Engineers of Japan",
number = "3",

}

TY - JOUR

T1 - Action detection of volleyball using features based on clustering of body trajectories

AU - Kubota, Eijiro

AU - Suzuki, Takahiro

AU - Honda, Masaaki

AU - Ikenaga, Takeshi

PY - 2016/1/1

Y1 - 2016/1/1

N2 - For creating new tactics of sports like volleyball, the analysis of player motion in real games becomes more and more important. However, since motion data needed for the analysis is captured by human observation currently, an automatic capturing system from video camera is highly expected to gather many useful data easily. This paper proposes an action detection algorithm of volleyball players using motion features based on clustering and aggregation of body trajectories. Since the body trajectories of arms and legs are similar, the clustering utilizes shape, location and density of their trajectories. Furthermore, the clustered feature values are aggregated by means of their mean and variance. Experimental results by using the motion detection system based on the proposed algorithm show that it averagely attains 0.9539 AUC of the ROC curve for the detection of four basic motions (block, receive, spike and toss) from the volleyball game video captured by high-definition cameras. This is 0.014775 higher than conventional methods.

AB - For creating new tactics of sports like volleyball, the analysis of player motion in real games becomes more and more important. However, since motion data needed for the analysis is captured by human observation currently, an automatic capturing system from video camera is highly expected to gather many useful data easily. This paper proposes an action detection algorithm of volleyball players using motion features based on clustering and aggregation of body trajectories. Since the body trajectories of arms and legs are similar, the clustering utilizes shape, location and density of their trajectories. Furthermore, the clustered feature values are aggregated by means of their mean and variance. Experimental results by using the motion detection system based on the proposed algorithm show that it averagely attains 0.9539 AUC of the ROC curve for the detection of four basic motions (block, receive, spike and toss) from the volleyball game video captured by high-definition cameras. This is 0.014775 higher than conventional methods.

KW - Clustering

KW - Dense trajectories

KW - Feature value aggregation

KW - Video detection

KW - Volleyball

UR - http://www.scopus.com/inward/record.url?scp=85041449326&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041449326&partnerID=8YFLogxK

U2 - 10.11371/iieej.45.373

DO - 10.11371/iieej.45.373

M3 - Article

AN - SCOPUS:85041449326

VL - 45

SP - 373

EP - 381

JO - Journal of the Institute of Image Electronics Engineers of Japan

JF - Journal of the Institute of Image Electronics Engineers of Japan

SN - 0285-9831

IS - 3

ER -