Study of Real Time Motion Object Detection for Integration of Virtual and Real World in Virtual Reality

• Main Authors: Hsiang-Lin Tsai, 蔡緗琳
• Other Authors: 朱中華
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/abt6y3

碩士 === 國立臺中科技大學 === 多媒體設計系碩士班 === 106 === In recent years, the development of virtual reality has become more mature, and its application in real life has become more diverse. This these is based on an AR smartphone that supports Google’s Project Tango with its two cameras - RGB camera and depth camera, to sense 3D real world. In this study, the average depth data of Point Clouds on the real 3D objects captured by the depth camera is used to determine whether there are 3D dynamic objects in the real scene. When the average depth data change exceeds the threshold of the average depth data, a 3D dynamic object can be detected. This threshold is also adopted to adjust 3D scanning speed and to update VR representation. We conduct detailed experimental analysis and verification, and completely analyze the physical error of the virtual and real integration in this thesis.