Результаты исследований: Публикации в книгах, отчётах, сборниках, трудах конференций › статья в сборнике материалов конференции › научная › Рецензирование
Position Tracking in 3D Space Based on a Data of a Single Camera. / Oleg, Iakushkin; Sevostyanov, Ruslan; Degtyarev, Alexander; Karpiy, P. E.; Kuzevanova, E. G.; Kitaeva, A. A.; Sergiev, S. A.
Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings. ред. / Beniamino Murgante; Osvaldo Gervasi; Elena Stankova; Vladimir Korkhov; Sanjay Misra; Carmelo Torre; Eufemia Tarantino; David Taniar; Ana Maria A.C. Rocha; Bernady O. Apduhan. Springer Nature, 2019. стр. 772-781 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Том 11622 LNCS).Результаты исследований: Публикации в книгах, отчётах, сборниках, трудах конференций › статья в сборнике материалов конференции › научная › Рецензирование
}
TY - GEN
T1 - Position Tracking in 3D Space Based on a Data of a Single Camera
AU - Oleg, Iakushkin
AU - Sevostyanov, Ruslan
AU - Degtyarev, Alexander
AU - Karpiy, P. E.
AU - Kuzevanova, E. G.
AU - Kitaeva, A. A.
AU - Sergiev, S. A.
N1 - Conference code: 19
PY - 2019/6/29
Y1 - 2019/6/29
N2 - 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.
AB - 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.
KW - Augmented reality
KW - Internet of things
KW - Optical flow
KW - Software
UR - http://www.scopus.com/inward/record.url?scp=85068614268&partnerID=8YFLogxK
UR - http://www.mendeley.com/research/position-tracking-3d-space-based-data-single-camera
U2 - 10.1007/978-3-030-24305-0_58
DO - 10.1007/978-3-030-24305-0_58
M3 - Conference contribution
AN - SCOPUS:85068614268
SN - 9783030243043
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 772
EP - 781
BT - Computational Science and Its Applications- ICCSA 2019 - 19th International Conference, Proceedings
A2 - Murgante, Beniamino
A2 - Gervasi, Osvaldo
A2 - Stankova, Elena
A2 - Korkhov, Vladimir
A2 - Misra, Sanjay
A2 - Torre, Carmelo
A2 - Tarantino, Eufemia
A2 - Taniar, David
A2 - Rocha, Ana Maria A.C.
A2 - Apduhan, Bernady O.
PB - Springer Nature
T2 - 19th International Conference on Computational Science and Its Applications, ICCSA 2019
Y2 - 1 July 2019 through 4 July 2019
ER -
ID: 43696607