Abstract
The objective of this study is the realization of recognition of human body movement using IMU (Inertial Measurement Unit). This paper shows segmentation and the recognition results of the movement of the right arm during Japanese drum performance to confirm the segmentation and recognition possibilities. In past studies the real-time recognition of human body movement using joint angles acquired by motion capture as been realized. The program of segmentation for joint angles is appropriately modified for angular velocity. And to adjust the parameter settings in order to match the segmentation results in manual segmentation results. The similar movement patterns are classified into each model. Recognition is performed using these models.
