Abstract

A high cost of equipment that solves the problem of tracking the position and direction of users in real time is one of factors that negatively affect the speed of development of the augmented reality industry. The urgency of this problem is a premise for the development of a financially available tracking system. In this research, we propose a software and hardware architecture of a system that solves three-dimensional tracking problems in a closed space and postures classification using neural network models. Distinctive feature of our system is the feasibility in borders of strictly limited computing power and the absence of any sensors placed on monitored objects. After setting the boundaries of the active area, all the necessary input data is provided by a static camera without an infrared filter. As an example of the implementation of a resource-limited solution, we present the assembly of this solution on a Raspberry Pi version 3 single board computer equipped with the Intel Neural Stick version 2 co-processor and a Raspberry version 2 NoIR camera. The first section of the article describes technical characteristics of the equipment used in the study. The second part is dedicated to the solution algorithm and its brief description. Further, in the third stage, the ways of data collection, necessary for a correct assessment of position, direction and posture are illustrated. The fourth, final section presents the results, discussion and possible directions for further work.

Original languageEnglish
Title of host publicationComputational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings
EditorsBeniamino Murgante, Osvaldo Gervasi, Elena Stankova, Vladimir Korkhov, Sanjay Misra, Carmelo Torre, Eufemia Tarantino, David Taniar, Ana Maria A.C. Rocha, Bernady O. Apduhan
PublisherSpringer Nature
Pages772-781
Number of pages10
ISBN (Print)9783030243043
DOIs
Publication statusPublished - 29 Jun 2019
Event19th International Conference on Computational Science and Its Applications, ICCSA 2019 - Saint Petersburg
Duration: 1 Jul 20194 Jul 2019

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11622 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference19th International Conference on Computational Science and Its Applications, ICCSA 2019
CountryRussian Federation
CitySaint Petersburg
Period1/07/194/07/19

Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Position Tracking in 3D Space Based on a Data of a Single Camera'. Together they form a unique fingerprint.

Cite this