Golf training in Mixed Reality

In this study, we propose a self sports learning system that gives multimodal feedback difference of motions in real time. Damp- ing DTW(Dynamic Time Warping), a proposed algorithm in this paper, allows the user to dynamically change the motion speed and repeat the certain motion without additional operation, therefore, the user can imitate the expert’s motion conveniently and accurately. The system provides two modes (training mode, replay mode) for learning a expert’s motion and the user can imitate the expert’s motion briefly through audio-visual feedback when training a motion, and after the training, the user can compare the difference of motion between themselves and the expert in detail. All feedback generated through the system is provided to the user by a mixed reality head-mounted display, therefore, the user can perceive the difference of the motion between a user and an expert in three dimensions, and can be immsered in the training.


Papers:

  • Yuka Tanaka, Homare Kon, Hideki Koike: A Real-Time Golf Swing Training System Using Sonification and Sound Image Localization, in Proc of ACM Virtual Reality Software and Technology (VRST 2018), article 66, 2018.
  • Atsuki Ikeda, Dong-Hyun Hwang, Hideki Koike, A Real-Time Projection System for Golf Training using Virtual Shadow, The First IEEE VR Workshop on Human Augmentation and Its Applications (HAA2019), co-located with IEEE VR 2019, 2019, to be published.
  • Atsuki Ikeda, Dong-Hyun Hwang, Hideki Koike, Self Sports Learning System with Multimodal Feedback using Domping Dynamic TIme Warping, in Proc. of International Conference on Artificial Reality and Telexistence and Eurographics Symposium on Virtual Environments (ICAT-EGVE), 2018. (download)