Abstract—The main aim of this paper is to propose the approach for visual estimation of selected parameters of the skeletal model. The outputs of this estimation form nodes of a human skeleton from a predefined skeletal model. The paper also analyzes mobile position sensors and also existing position sensing solutions based on a visual approach. This analysis introduces the physical principle of mobile device sensors and their use in practice. It also introduces algorithms and existing position sensing solutions. Experimental implementation is focused on the functionality of the designed software solution, tested for Android operating system. Testing was carried out in the laboratory environment and focused on the accuracy of height measurement and angular measurement. The accuracy of the measurement using different devices is evaluated and the output of the solution utilizes standardized novel human body description format.
Index Terms—Capturing position, skeletal model, ARToolkit, android.
The authors are with the Department of Computers and Informatics, Technical University of Košice, Kosice, Slovakia (e-mail: ondrej.kainz@cnl.sk, filip.chovanec@student.tuke.sk, miroslav.michalko@cnl.sk, frantisek.jakab@cnl.sk).
[PDF]
Cite: O. Kainz, F. Chovanec, M. Michalko, and F. Jakab, "Mobile Solution for Estimating the Movement of Device," International Journal of Future Computer and Communication vol. 7, no. 1, pp. 1-5, 2018.
