主催: 一般社団法人 日本機械学会
会議名: スポーツ工学・ヒューマンダイナミクス2019
開催日: 2019/10/25 - 2019/10/27
The objective of our research is to develop the pose estimation algorithm for Karate motions using RGB Camera. Kata is the representation of Karate’s self-defense techniques strung together into a performance routine. Kata is judged based on several technical and physical criteria; including speed and strength. For this reason, quality evaluation for Karate motions by technology is difficult. In this research, we created a data set targeting karate movement. We also developed the pose estimation algorithm for Karate motions by deep learning using this data set.