Development of Game Module Using Technology of Human Pose Estimation for the Neurological Rehabilitation System
- Authors: Pavlikov A.E.1
-
Affiliations:
- Moscow Technical University of Communications and Informatics
- Issue: Vol 12, No 1 (2025)
- Pages: 116-128
- Section: INFORMATICS AND INFORMATION PROCESSING
- URL: https://journals.eco-vector.com/2313-223X/article/view/679152
- DOI: https://doi.org/10.33693/2313-223X-2025-12-1-116-128
- EDN: https://elibrary.ru/MQMTIG
- ID: 679152
Cite item
Open Access
Access granted
Abstract
The development of deep learning algorithms makes it possible to extend the scope of their application to various spheres of human life. Today, deep neural networks can solve problems in natural language processing, data generation, computer vision and so on. In this paper, a game module for a neurological rehabilitation system using Human pose estimation algorithm on video is designed and implemented. Different HPE algorithms including REMOTE, MAPN and MediaPipe Pose were considered in the research process and their comparative analysis on PCK, FPS and MAP metrics was done. As a result, MediaPipe Pose was selected to provide the best balance between accuracy and performance. The developed game module allows patients to perform movements in an interactive environment, and doctors to track rehabilitation progress based on movement parameters such as number of executions, time between executions, number of execution errors, and types of errors. The module allows doctors to select a difficulty level for the current game session to work with patients at different stages of rehabilitation.
Full Text
About the authors
Artyom E. Pavlikov
Moscow Technical University of Communications and Informatics
Author for correspondence.
Email: a.e.pavlikov@mtuci.ru
ORCID iD: 0009-0001-6165-7474
SPIN-code: 7266-2752
Scopus Author ID: 58204705000
Assistant of the Department of PI
Russian Federation, MoscowReferences
- Hak Gu Kim, Sangmin Lee, Seongyeop Kim. Towards a better understanding of VR sickness: Physical symptom prediction for VR contents. arXiv. 2021.
- Steinmetz J.D., Seeher K.M., Schiess N. et al. Global, regional, and national burden of disorders affecting the nervous system, 1990–2021: A systematic analysis for the Global Burden of Disease Study 2021. The Lancet Neurology. 2024. Vol. 23. Issue 4. Pp. 344–381.
- Sungtaek Cho, Dongyeon Kim, Sungon Lee. A comparative evaluation of a single and stereo lighthouse systems for 3-D estimation. IEEE Sensors Journal. 2021. P. 99.
- Xianzheng Ma, Hossein Rahmani, Zhipeng Fan et al. REMOTE: Reinforced motion transformation network for semi-supervised 2D pose estimation in videos. In: The Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI-22). 2022. Pp. 1944–1952.
- Zhipeng Fan, Jun Liu, Yao Wang. Motion adaptive pose estimation from compressed videos. IEEE/CVF International Conference on Computer Vision (ICCV). 2021. Pp. 11719–11728.
- Bychkov A., Kiseleva T., Maslova E. Usage of convolutional neural networks for image classification. Bulletin of the Siberian State Industrial University. 2023. No. 1. Pp. 39–49. (In Rus.)
- Karyakin A.V. Research of the problem of human detection using computer vision. URL: https://www.researchgate.net/publication/381037033_Issledovanie_zadaci_detektirovania_celoveka_s_pomosu_komputernogo_zrenia (data of accesses: 24.04.2025)
- Kislenko S.L., Menzhega M.M. Use of modern technical means in the process of fixing the crime scene examination’s results. Bulletin of the Institute of Law of Bashkir State University. 2024. No. 7(3(23)). Pp. 108–123. (In Rus.)
- Konovalov A.N., Pilipenko Yu.V. et. al. Augmented reality as a method of neuronavigation for extra-intracranial microanastomosis. Russian Journal of Operative Surgery and Clinical Anatomy. 2024. Vol. 8. No. 3. Pp. 28–34. (In Rus.)
- Kudinov Ya.O. Study of the possibility of classifying paintings using computer vision. URL: https://www.researchgate.net/publication/377219522_Klassifikacia_kartin_s_pomosu_komputernogo_zrenia (data of accesses: 30.01.2025).
- Leonov I.Yu. Human pose estimation on images of yoga asanas. URL: https://www.researchgate.net/publication/381116740_Human_pose_estimation_na_izobrazeniah_asan_v_joge (data of accesses: 30.01.2025)
- Pavlikov A.E., Gorodnichev M.G. Overview of technologies for determining the position of the human body. Models, systems, networks in economics, technology, nature and society. 2023. No. 3. Pp. 81–97. (In Rus.)
- Pisar N.V. Virtual and augmented reality technologies as a tool for teaching communication in Russian. Prepodavatel XXI vek. Russian Journal of Education. 2023. No. 3. Part 1. Pp. 212–222. (In Rus.)
- Khuako V.O., Abakumov A.A. Determination of human body position using neural networks. URL: https://www.researchgate.net/publication/380785157_Opredelenie_polozenia_tela_celoveka_s_ispolzovaniem_nejronnyh_setej (data of accesses: 30.01.2025).
Supplementary files
