Abstract
To provide suitable care for concussion, objective and timely detection of high-risk event is crucial. Currently it depends on monitoring by medical doctors, and there is a certain risk of missing high-risk events. A few attempts introducing video analysis have been reported, but those approaches require labeling by experts, which is skill-dependent, and time and cost consuming. To achieve objective detection of high-risk tackle without human intervention, 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 tackles and 155 high-risk tackles were extracted. Poses of tackler and ball carrier were estimated by deep learning, then were evaluated by machine learning. The proposed method resulted AUC 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 of concussion. This result indicates that our approach combining deep learning and machine learning opens the way for objective and timely detection of high-risk events in rugby and other contact sports.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Among several variants of Rugby, here we focus on Rugby Union, played between two teams of 15 players each.
- 2.
Tackle type: {Active shoulder tackle, Passive shoulder tackle, Smother tackle, Tap tackle, Lift tackle, No arm tackle, High tackle}, Tackle direction: {Front-on, On angle, Side-on, Back}, Accelerating player: {Tackler, Ball carrier, Both, Neither}, Tackler and BC speed: {High, Low, Stationary}, Tackler and BC body positions: {Upright, Bent at the waist, Falling/diving}.
References
Altmann, A.: Permutation importance: a corrected feature importance measure. Bioinformatics 26, 1340–1347 (2010)
Bridgeman, L.: Multi-person 3D pose estimation and tracking in sports. In: CVPRW (2019)
Cai, Z.: Temporal hockey action recognition via pose and optical flows. In: CVPRW (2019)
Einfalt, M.: Decoupling video and human motion: towards practical event detection in athlete recordings. In: CVPRW (2020)
Fuller, C.W.: Evaluation of World Rugby’s concussion management process: results from Rugby World Cup 2015. Br. J. Sports Med. 51, 64–69 (2017)
Gardner, A.J.: A systematic review and meta-analysis of concussion in rugby union. Sports Med. 44, 1717–1731 (2014)
Hopkinson, M.: Rugby league ball carrier injuries: the relative importance of tackle characteristics during the European Super League. Eur. J. Sport Sci. 22, 269–278 (2021)
Kulkarni, K.M.: Table tennis stroke recognition using two-dimensional human pose estimation. In: CVPRW (2021)
Lin, T.-Y., et al.: Microsoft COCO: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8693, pp. 740–755. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10602-1_48
Morissette, M.P.: Associations between concussion and risk of diagnosis of psychological and neurological disorders: a retrospective population-based cohort study. Family Med. Commun. Health (2020)
Nekoui, M.: FALCONS: FAst learner-grader for CONtorted poses in sports. In: CVPRW (2020)
Ogata, R.: Temporal distance matrices for squat classification. In: CVPRW (2019)
Piergiovanni, A.J.: Early detection of injuries in MLB pitchers from video. In: CVPRW (2019)
Ramana Murthy, O.V.: Injury mechanism classification in soccer videos. In: ICCVW (2015)
Takahashi, K.: Human pose as calibration pattern: 3D human pose estimation with multiple unsynchronized and un-calibrated cameras. In: CVPRW (2018)
Tucker, R.: Risk factors for head injury events in professional rugby union: a video analysis of 464 head injury events to inform proposed injury prevention strategies. Br. J. Sports Med. (2017)
Vincent, M.: Tackler and ball-carrier technique during moderate and severe injuries (\(\ge \)8 days lost) compared with player-matched and team-matched injury-free controls in Elite Rugby Union. Br. J. Sports Med. (2021)
Williams, S.: A meta-analysis of injuries in senior men’s professional rugby union. Sports Med. (2013)
Xingyi, Z.: Tracking objects as points. arXiv (2020)
Zecha, D.: Kinematic pose rectification for performance analysis and retrieval in sports. In: CVPRW (2018)
Zecga, D.: Refining joint locations for human pose tracking in sports videos. In: CVPRW (2019)
Zimmerman, K.A.: White matter abnormalities in active elite adult rugby players. Brain Commun. (2021)
Acknowledgement
The authors would like to thank Japan Rugby Football Union for their support. We also thank J Sports Corporation for providing video data. This work was supported by MEXT “Innovation Platform for Society 5.0” Program Grant Number JP-MXP0518071489.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Nishio, M. et al. (2023). Objective Detection of High-Risk Tackle in Rugby by Combination of Pose Estimation and Machine Learning. In: Takama, Y., Yada, K., Satoh, K., Arai, S. (eds) New Frontiers in Artificial Intelligence. JSAI-isAI 2022. Lecture Notes in Computer Science(), vol 13859. Springer, Cham. https://doi.org/10.1007/978-3-031-29168-5_15
Download citation
DOI: https://doi.org/10.1007/978-3-031-29168-5_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-29167-8
Online ISBN: 978-3-031-29168-5
eBook Packages: Computer ScienceComputer Science (R0)