7. Interactive Projection Mapping using Human Detection by Machine Learning

Makoto Sakamoto1, Takahiro Shinoda1, Kenji Sakoma1, Takahiro Ishizu1, Amane Takei1, Takao Ito2
1Department of Computer Science and System Engineering, University of Miyazaki, Miyazaki-City, 889-2192, Japan
2Graduate School of Engineering, Hiroshima University, Higashi-Hiroshima, 739-8527, Japan
pp. 86-90
ABSTRACT
In recent years, “Entertainment Computing” (EC) has attracted attention and has become one of the major industries in Japan. “Projection Mapping” is well known in this EC. Projection mapping is a video technology that creates a new space by synthesizing space and video using a projector. Among them, many people are fascinated by works that create a fantastic world by combining dancer performance and projection mapping. However, these works require the performer to accurately align with the coordinates of the image objects in the projection mapping, which is not easy for everyone. In this study, we aim to entertain not only the people who see the projection mapping but also performers. Therefore, we have prototyped an interactive projection mapping that changes according to user movement. This time, we focused on sports and projected the ball to the user to experience baseball pitching and soccer lifting. Furthermore, we conducted a questionnaire survey to evaluate the sense of use of this system, and the results showed that many people can enjoy by this projection mapping

ARTICLE INFO
Article History
Received 25 October 2019
Accepted 03 July 2020

Keywords
Entertainment computing
EC
Projection mapping
Kinect

JAALR1207

Download article(PDF)