Research on automatic generation of sense of motion based on video

-2nd Report: Representing skill differences in dance movements by generating haptic vibrations-

Abstract

We propose a method to generate vibration sensations automatically following the motion of a person in a video so that anyone can create content with motion sensations easily. We have already confirmed that the impact and presence of the videos increased when vibration stimuli generated from sound effects were added to the jumping motion. In this study, we propose a method of presenting a sensation linked to a performer’s movement using vibration stimuli to express differences in skill in movement. The method uses human pose estimation to detect the motion of a person in a video. Then, it generates vibration waveforms based on the perceived intensity proportional to motion. Subject experiments verify the effectiveness of the proposed method.