Constrained multi-target tracking for team sports activities

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

In sports analysis, player tracking is essential to the extraction of statistics such as speed, distance and direction of motion. Simultaneous tracking of multiple people is still a very challenging computer vision problem to which there is no satisfactory solution. This is especially true for sports activities, for which people often wear similar uniforms, move quickly and erratically, and have close interactions with each other. In this paper, we introduce a multi-target tracking algorithm suitable for team sports activities. We extend an existing algorithm by including an automatic estimation of the occupancy of the observed field and the duration of stable periods without people entering or leaving the field. This information is included as a constraint to the existing offline tracking algorithm in order to construct more reliable trajectories. On data from two challenging sports scenarios—an indoor soccer game captured with thermal cameras and an outdoor soccer training session captured with RGB camera—we show that the tracking performance is improved on all sequences. Compared to the original offline tracking algorithm, we obtain improvements of 3–7% in accuracy. Furthermore, the method outperforms two state-of-the-art trackers.
Close

Details

In sports analysis, player tracking is essential to the extraction of statistics such as speed, distance and direction of motion. Simultaneous tracking of multiple people is still a very challenging computer vision problem to which there is no satisfactory solution. This is especially true for sports activities, for which people often wear similar uniforms, move quickly and erratically, and have close interactions with each other. In this paper, we introduce a multi-target tracking algorithm suitable for team sports activities. We extend an existing algorithm by including an automatic estimation of the occupancy of the observed field and the duration of stable periods without people entering or leaving the field. This information is included as a constraint to the existing offline tracking algorithm in order to construct more reliable trajectories. On data from two challenging sports scenarios—an indoor soccer game captured with thermal cameras and an outdoor soccer training session captured with RGB camera—we show that the tracking performance is improved on all sequences. Compared to the original offline tracking algorithm, we obtain improvements of 3–7% in accuracy. Furthermore, the method outperforms two state-of-the-art trackers.
Original languageEnglish
Article number2
JournalIPSJ Transactions on Computer Vision and Applications
Volume10
Issue number1
Number of pages11
ISSN1882-6695
DOI
Publication statusPublished - Jan 2018
Publication categoryResearch
Peer-reviewedYes

    Research areas

  • Counting people, Soccer, Sports analysis, Tracking people
ID: 267849291