License: Creative Commons Attribution 4.0 International license (CC BY 4.0)
When quoting this document, please refer to the following
DOI: 10.4230/DagSemProc.08372.5
URN: urn:nbn:de:0030-drops-16891
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2008/1689/
Go to the corresponding Portal

Pers, Janez ; Kristan, Matej ; Perse, Matej ; Kovacic, Stanislav

Analysis of Player Motion in Sport Matches

pdf-format:
08372.PersJanez.Paper.1689.pdf (0.8 MB)

Abstract

The system for analysis of player motion during sport matches, developed at University of Ljubljana will be presented. The system allows for non-intrusive measurement of positions of all players in indoor sports through whole match using only inexpensive video equipment - cameras mounted on the ceiling of the sports hall. Tracking process (obtaining trajectories from videos) is automatic and only supervised by operator, to initialize player positions at the beginning and correct the mistakes during the tracking. The software provides means for user friendly calibration of video data - using court markings of each supported sport (e.g. european handball, basketball, squash, tennis...) as reference coordinates. Manual annotations can be added, to complement the quantitative data. Software keeps synchronization between annotations and trajectory data and provides means to use custom annotation dictionaries. Due to calibration, the results are provided in court coordinates (meters, centimeters) and can be exported (synchronized with annotations in same file) for further analysis with any application (e.g. excel, SPSS). Software itself supports several kinds of graphical data presentations.

If time allows, the software itself will be demonstrated with examples from different sports.

BibTeX - Entry

@InProceedings{pers_et_al:DagSemProc.08372.5,
  author =	{Pers, Janez and Kristan, Matej and Perse, Matej and Kovacic, Stanislav},
  title =	{{Analysis of Player Motion in Sport Matches}},
  booktitle =	{Computer Science in Sport - Mission and Methods},
  pages =	{1--10},
  series =	{Dagstuhl Seminar Proceedings (DagSemProc)},
  ISSN =	{1862-4405},
  year =	{2008},
  volume =	{8372},
  editor =	{Arnold Baca and Martin Lames and Keith Lyons and Bernhard Nebel and Josef Wiemeyer},
  publisher =	{Schloss Dagstuhl -- Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2008/1689},
  URN =		{urn:nbn:de:0030-drops-16891},
  doi =		{10.4230/DagSemProc.08372.5},
  annote =	{Keywords: Motion analysis, computer vision, performance analysis}
}

Keywords: Motion analysis, computer vision, performance analysis
Collection: 08372 - Computer Science in Sport - Mission and Methods
Issue Date: 2008
Date of publication: 20.11.2008

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI