Real-time body motion coding for skills teaching

2018 Fiscal Year Final Research Report

• Project/Area Number: 15K16257
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Educational technology
• Research Institution: Mie University
• Principal Investigator: Sakamoto Ryota 三重大学, 医学部附属病院, 助教 (10581879)
• Project Period (FY): 2015-04-01 – 2019-03-31
• Keywords: モーションキャプチャー / 動作教示 / 動作計測 / バイオメカニクス / 健康・福祉工学

Outline of Final Research Achievements

This project aims at finding out notable features of body motions in order to teach a skill of a sport, a dance, or a medical procedure. Motion data of the body acquired with a motion capture manufactured building is a tremendous amount to each joint of the whole body. Therefore, coding methods that reducing unwanted data while keeping the potential features are desired. The author examined joints contributing to a characteristic posture such as foot plant and top position in pitching motion in baseball. The author also proposed a method to find the difference between novices and experts by machine learning in a medical operation.

Free Research Field

教育工学

Academic Significance and Societal Importance of the Research Achievements

教科書等で自習できるような知識の習得と異なり，スポーツや動作に基づく技能の学習は指導者が直接教示しなければならいないことが多い．これを自学しようとすれば，学習者自身の感覚で課題点を見出さなければならない．ビデオにて記録された手本，あるいは自身の映像を学習者が視聴する形での自学でも学習者の主観に基づく点で同様である．申請者はこの要因として考えられる，対象の動作における正解といえるものが何かを定めること，またそれを指導するために伝えることを解決するための研究を実施した．