JSAI2022

Presentation information

International Session

International Session » ES-2 Machine learning

[1S5-IS-2a] Machine learning

Tue. Jun 14, 2022 4:20 PM - 6:00 PM Room S (Online S)

Chair: Toshihiko Matsuka (Chiba University)

5:40 PM - 6:00 PM

[1S5-IS-2a-05] Objective detection of high-risk tackle in rugby by combination of pose estimation and machine learning

〇Monami Nishio1, Naoki Nonaka1, Ryo Fujihira1, Hidetaka Murakami2, Takuya Tajima3, Mutsuo Yamada4, Akira Maeda5,6, Jun Seita1 (1. Advanced Data Science Project, RIKEN Information R&D and Strategy Headquarters, 2. Murakami Surgical Hospital, 3. Division of Orthopaedic Surgery, Department of Sensory and Motor Organs, Faculty of Medicine, University of Miyazaki, 4. Faculty of Health and Sport Sciences, Ryutsu Keizai University, 5. Hakata Knee & Sports Clinic, 6. Department of Sports Medicine and Science, Faculty of Human Health, Kurume University)

Regular

Keywords:machine learning, pose estimation, sports

To provide suitable care for concussion, objective and timely detection of high-risk tackle is crucial in the field of contact sports, such as rugby. Currently it depends on monitoring by match officials, and there is a certain risk of missing high-risk events. A few attemps introducing video analysis have been reported, but those approaches require labeling by experts, which is skill-dependent, and also time and cost consuming.
To achieve objective and timely detection of high-risk tackle, we developed a method combining pose estimation by deep-learning and pose evaluation by machine learning. From match videos of Japan Rugby Top League in 2016~2018 seasons, 238 low-risk tackle and 155 high-risk tackle were extracted. Poses of tackler and ball carrier were estimated by deep learning, then were evaluated by machine learning.
The proposed method resulted AUROC-score 0.85 and outperformed the previously reported rule-based method. Also, the features extracted by the machine learning model, such as upright positions of tackler/ball carrier, tackler's arm dropped in extended position, were consistent with the known risk factors.
This result indicates that our approach combining deep-learning and machine learning opens the way for objective and real-time detection of high-risk tackle in rugby and other contact sports.

Authentication for paper PDF access

A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.

Password