ellibsЭлектронные библиотекиRussian Digital Libraries Journal1562-5419Казанский (Приволжский) федеральный университет10.26907/1562-5419-2025-28-3-573-600ellibs-584Research ArticleСтатьиСравнительный анализ библиотек для детектирования позы человека в условиях работы на мобильных устройствахComparative Analysis of Libraries for Human Pose Detection in Mobile Device EnvironmentsЯркоЕгор ИгоревичYarkoEgor Igorevichyarkoeg@gmail.comКазанский (Приволжский) федеральный университетKazan (Volga region) Federal University202523062025283573600Copyright © Ярко Е.И., 20252025Ярко Е.И.Yarko E.I.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://ellibs.elpub.ru/jour/article/view/584

   Оценка положения тела человека (Human Pose Estimation, HPE) стала одной из наиболее актуальных тем в исследованиях в области компьютерного зрения. Эта технология может применяться в различных сферах, таких как видеонаблюдение, медицинская помощь и анализ спортивных движений. В связи с растущим спросом на HPE за последние 20 лет было разработано большое количество библиотек для этой технологии. C 2017 года опубликовано множество алгоритмов HPE, основанных на скелетной модели, которые были упакованы в библиотеки для удобства использования исследователями. Эти библиотеки важны для исследователей, которые хотят интегрировать их в реальные приложения для видеонаблюдения, медицинской помощи и анализа спортивных движений. В работе рассмотрены преимущества и недостатки четырёх популярных передовых библиотек HPE для распознавания поз человека, которые могут работать на мобильных устройства: Lightweight OpenPose, PoseNet, MoveNet и Blase Pose.

Human Pose Estimation (HPE) has become one of the most relevant topics in computer vision research. This technology can be applied in various fields such as video surveillance, medical care, and sports motion analysis. Due to the increasing demand for HPE, many libraries for this technology have been developed in the last 20 years. Since 2017, many HPE algorithms based on skeletal model have been published and packaged into libraries for easy use by researchers. These libraries are important for researchers who want to integrate them into real-world applications for video surveillance, medical care, and sports motion analysis. This paper investigates the strengths and weaknesses of four popular HPE advanced human pose recognition libraries that can run on mobile devices: Lightweight OpenPose, PoseNet, MoveNet, and Blase Pose.

поза человекаHuman Pose EstimationHPEдетектирование позыкомпьютерное зрениемобильные устройствадополненная реальностьLightweight OpenPosePoseNetMoveNetBlazePoseскелетная модельHuman poseHuman Pose EstimationHPEpose detectioncomputer visionmobile devicesaugmented realityLightweight OpenPosePoseNetMoveNetBlazePoseskeletal modelReferencesSu C., Li J.; Zhang S., Xing J., Gao W., Tian Q. Pose-driven deep convolutional model for person re-identification // Proceedings of the IEEE International Conference on Computer Vision. 2017. P. 3960–3969.Su C., Li J.; Zhang S., Xing J., Gao W., Tian Q. Pose-driven deep convolutional model for person re-identification // Proceedings of the IEEE International Conference on Computer Vision. 2017. P. 3960–3969.Xu J., Zhao R., Zhu F., Wang H., Ouyang W. Attention-aware compositional network for person re-identification // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018. P. 2119–2128.Xu J., Zhao R., Zhu F., Wang H., Ouyang W. Attention-aware compositional network for person re-identification // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2018. P. 2119–2128.Thyagarajmurthy A., Ninad M.G., Rakesh B.G., Niranjan S., Manvi B. Anomaly detection in surveillance video using pose estimation // Proceedings of the Emerging Research in Electronics, Computer Science and Technology; Springer. 2019. P. 753–766.Thyagarajmurthy A., Ninad M.G., Rakesh B.G., Niranjan S., Manvi B. Anomaly detection in surveillance video using pose estimation // Proceedings of the Emerging Research in Electronics, Computer Science and Technology; Springer. 2019. P. 753–766.Lamas A., Tabik S., Montes A.C., Pérez-Hernández F., García J., Olmos R., Herrera F. Human pose estimation for mitigating false negatives in weapon detection in video-surveillance // Neurocomputing. 2022. Vol. 489. P. 488–503.Lamas A., Tabik S., Montes A.C., Pérez-Hernández F., García J., Olmos R., Herrera F. Human pose estimation for mitigating false negatives in weapon detection in video-surveillance // Neurocomputing. 2022. Vol. 489. P. 488–503.Yoo H.R., Lee B.H. An OpenPose-based child abuse decision system using surveillance video // J. Korea Inst. Inf. Commun. Eng. 2019. Vol. 23. P. 282–290.Yoo H.R., Lee B.H. An OpenPose-based child abuse decision system using surveillance video // J. Korea Inst. Inf. Commun. Eng. 2019. Vol. 23. P. 282–290.Park J.H., Song K., Kim Y.-S. A Kidnapping Detection Using Human Pose Estimation in Intelligent Video Surveillance Systems // J. Korea Soc. Comput. Inf. 2018. Vol. 23. P. 9–16.Park J.H., Song K., Kim Y.-S. A Kidnapping Detection Using Human Pose Estimation in Intelligent Video Surveillance Systems // J. Korea Soc. Comput. Inf. 2018. Vol. 23. P. 9–16.Chang Y.J., Chen S.F., Huang J.D. A Kinect-based system for physical rehabilitation: A pilot study for young adults with motor disabilities // Res. Dev. Disabil. 2011. Vol. 32. P. 2566–2570.Chang Y.J., Chen S.F., Huang J.D. A Kinect-based system for physical rehabilitation: A pilot study for young adults with motor disabilities // Res. Dev. Disabil. 2011. Vol. 32. P. 2566–2570.Hassan H.A., Abdallah B.H., Abdallah A.A., Abdel-Aal R.O., Numan R.R., Darwish A.K., El-Behaidy W.H. Automatic Feedback for Physiotherapy Exercises Based On PoseNet // FCAI-Inform. Bull. 2020. Vol. 2. P. 10–14.Hassan H.A., Abdallah B.H., Abdallah A.A., Abdel-Aal R.O., Numan R.R., Darwish A.K., El-Behaidy W.H. Automatic Feedback for Physiotherapy Exercises Based On PoseNet // FCAI-Inform. Bull. 2020. Vol. 2. P. 10–14.Shapoval S., García Zapirain B., Mendez Zorrilla A., Mugueta-Aguinaga I. Biofeedback applied to interactive serious games to monitor frailty in an elderly population // Appl. Sci. 2021. Vol. 11. P. 3502.Shapoval S., García Zapirain B., Mendez Zorrilla A., Mugueta-Aguinaga I. Biofeedback applied to interactive serious games to monitor frailty in an elderly population // Appl. Sci. 2021. Vol. 11. P. 3502.Chua J., Ong L.Y., Leow M.C. Telehealth using PoseNet-based system for in-home rehabilitation // Future Internet. 2021. Vol. 13. P. 173.Chua J., Ong L.Y., Leow M.C. Telehealth using PoseNet-based system for in-home rehabilitation // Future Internet. 2021. Vol. 13. P. 173.Kim W., Sung J., Saakes D., Huang C., Xiong S. Ergonomic postural assessment using a new open-source human pose estimation technology (OpenPose) // Int. J. Ind. Ergon. 2021. Vol. 84. P. 103164.Kim W., Sung J., Saakes D., Huang C., Xiong S. Ergonomic postural assessment using a new open-source human pose estimation technology (OpenPose) // Int. J. Ind. Ergon. 2021. Vol. 84. P. 103164.Jawale C.D., Joshi K.A., Gogate S.K., Badgujar C. Elcare: Elderly Care with Fall Detection // J. Phys. Conf. Ser. 2022. Vol. 2273. P. 012019.Jawale C.D., Joshi K.A., Gogate S.K., Badgujar C. Elcare: Elderly Care with Fall Detection // J. Phys. Conf. Ser. 2022. Vol. 2273. P. 012019.Kapoor R., Jaiswal A., Makedon F. Light-weight seated posture guidance system with machine learning and computer vision // Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive Environments. 2022. P. 595–600.Kapoor R., Jaiswal A., Makedon F. Light-weight seated posture guidance system with machine learning and computer vision // Proceedings of the 15th International Conference on PErvasive Technologies Related to Assistive Environments. 2022. P. 595–600.Landi H. Google, ProMedica Team up with Include Health to Tap into Growing Virtual MSK Market // Fierce Healthcare. URL: https://www.fiercehealthcare.com/tech/google-promedica-team-up-includehealth-to-tap-into-virtual-msk-marketLandi H. Google, ProMedica Team up with Include Health to Tap into Growing Virtual MSK Market // Fierce Healthcare. URL: https://www.fiercehealthcare.com/tech/google-promedica-team-up-includehealth-to-tap-into-virtual-msk-marketChen W., Jiang Z., Guo H., Ni X. Fall detection based on key points of human-skeleton using OpenPose // Symmetry. 2020. Vol. 12. P. 744.Chen W., Jiang Z., Guo H., Ni X. Fall detection based on key points of human-skeleton using OpenPose // Symmetry. 2020. Vol. 12. P. 744.Zou J., Li B., Wang L., Li Y., Li X., Lei R., Sun S. Intelligent fitness trainer system based on human pose estimation // Proceedings of the International Conference On Signal and Information Processing, Networking and Computers. 2018. P. 593–599.Zou J., Li B., Wang L., Li Y., Li X., Lei R., Sun S. Intelligent fitness trainer system based on human pose estimation // Proceedings of the International Conference On Signal and Information Processing, Networking and Computers. 2018. P. 593–599.Suda S., Makino Y., Shinoda H. Prediction of volleyball trajectory using skeletal motions of setter player // Proceedings of the 10th Augmented Human International Conference. 2019. P. 1–8.Suda S., Makino Y., Shinoda H. Prediction of volleyball trajectory using skeletal motions of setter player // Proceedings of the 10th Augmented Human International Conference. 2019. P. 1–8.Wang J., Qiu K., Peng H., Fu J., Zhu J. AI coach: Deep human pose estimation and analysis for personalized athletic training assistance // Proceedings of the 27th ACM International Conference on Multimedia. 2019. P. 374–382.Wang J., Qiu K., Peng H., Fu J., Zhu J. AI coach: Deep human pose estimation and analysis for personalized athletic training assistance // Proceedings of the 27th ACM International Conference on Multimedia. 2019. P. 374–382.Jeon H., Yoon Y., Kim D. Lightweight 2D human pose estimation for fitness coaching system // Proceedings of the 2021 36th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC). 2021. P. 1–4.Jeon H., Yoon Y., Kim D. Lightweight 2D human pose estimation for fitness coaching system // Proceedings of the 2021 36th International Technical Conference on Circuits/Systems, Computers and Communications (ITC-CSCC). 2021. P. 1–4.Li Y.C., Chang C.T., Cheng C.C., Huang Y.L. Baseball Swing Pose Estimation Using OpenPose // Proceedings of the 2021 IEEE International Conference on Robotics, Automation and Artificial Intelligence (RAAI). 2021. P. 6–9.Li Y.C., Chang C.T., Cheng C.C., Huang Y.L. Baseball Swing Pose Estimation Using OpenPose // Proceedings of the 2021 IEEE International Conference on Robotics, Automation and Artificial Intelligence (RAAI). 2021. P. 6–9.Park H.J., Baek J.W., Kim J.H. Imagery based Parametric Classification of Correct and Incorrect Motion for Push-up Counter Using OpenPose // Proceedings of the 2020 IEEE 16th International Conference on Automation Science and Engineering (CASE). 2020. P. 1389–1394.Park H.J., Baek J.W., Kim J.H. Imagery based Parametric Classification of Correct and Incorrect Motion for Push-up Counter Using OpenPose // Proceedings of the 2020 IEEE 16th International Conference on Automation Science and Engineering (CASE). 2020. P. 1389–1394.Nguyen H.T.P., Woo Y., Huynh N.N., Jeong H. Scoring of Human Body-Balance Ability on Wobble Board Based on the Geometric Solution // Appl. Sci. 2022. Vol. 12. P. 5967.Nguyen H.T.P., Woo Y., Huynh N.N., Jeong H. Scoring of Human Body-Balance Ability on Wobble Board Based on the Geometric Solution // Appl. Sci. 2022. Vol. 12. P. 5967.Patil A., Rao D., Utturwar K., Shelke T., Sarda E. Body Posture Detection and Motion Tracking using AI for Medical Exercises and Recommendation System // ITM Web Conf. 2022. Vol. 44. P. 03043.Patil A., Rao D., Utturwar K., Shelke T., Sarda E. Body Posture Detection and Motion Tracking using AI for Medical Exercises and Recommendation System // ITM Web Conf. 2022. Vol. 44. P. 03043.Devanandan M., Rasaratnam V., Anbalagan M.K., Asokan N., Panchendrarajan R., Tharmaseelan J. Cricket Shot Image Classification Using Random Forest // Proceedings of the 2021 3rd International Conference on Advancements in Computing (ICAC). 2021. P. 425–430.Devanandan M., Rasaratnam V., Anbalagan M.K., Asokan N., Panchendrarajan R., Tharmaseelan J. Cricket Shot Image Classification Using Random Forest // Proceedings of the 2021 3rd International Conference on Advancements in Computing (ICAC). 2021. P. 425–430.Joseph R., Ayyappan M., Shetty T., Gaonkar G., Nagpal A. BeFit—A Real-Time Workout Analyzer // Proceedings of the Sentimental Analysis and Deep Learning. 2022. P. 303–318.Joseph R., Ayyappan M., Shetty T., Gaonkar G., Nagpal A. BeFit—A Real-Time Workout Analyzer // Proceedings of the Sentimental Analysis and Deep Learning. 2022. P. 303–318.Mahendran N. Deep Learning for Fitness // arXiv. 2021. arXiv:2109.01376.Mahendran N. Deep Learning for Fitness // arXiv. 2021. arXiv:2109.01376.Agarwal S., Gupta M., Khandelwal S., Jain P., Aggarwal A., Singh D., Mishra V.K. FitMe: A Fitness Application for Accurate Pose Estimation Using Deep Learning // Proceedings of the 2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC). 2021. P. 232–237.Agarwal S., Gupta M., Khandelwal S., Jain P., Aggarwal A., Singh D., Mishra V.K. FitMe: A Fitness Application for Accurate Pose Estimation Using Deep Learning // Proceedings of the 2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC). 2021. P. 232–237.Nakai M., Tsunoda Y., Hayashi H., Murakoshi H. Prediction of basketball free throw shooting by OpenPose // Proceedings of the JSAI International Symposium on Artificial Intelligence. 2018. P. 435–446.Nakai M., Tsunoda Y., Hayashi H., Murakoshi H. Prediction of basketball free throw shooting by OpenPose // Proceedings of the JSAI International Symposium on Artificial Intelligence. 2018. P. 435–446.Zheng C., Wu W., Chen C., Yang T., Zhu S., Shen J., Kehtarnavaz N., Shah M. Deep learning-based human pose estimation: A survey // arXiv. 2020, arXiv:2012.13392.Zheng C., Wu W., Chen C., Yang T., Zhu S., Shen J., Kehtarnavaz N., Shah M. Deep learning-based human pose estimation: A survey // arXiv. 2020, arXiv:2012.13392.Papandreou G., Zhu T., Kanazawa N., Toshev A., Tompson J., Bregler C., Murphy K. Towards accurate multi-person pose estimation in the wild // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu. 2017. P. 4903–4911.Papandreou G., Zhu T., Kanazawa N., Toshev A., Tompson J., Bregler C., Murphy K. Towards accurate multi-person pose estimation in the wild // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu. 2017. P. 4903–4911.Jo B., Kim S. Comparative Analysis of OpenPose, PoseNet, and MoveNet Models for Pose Estimation in Mobile Devices // Trait. du Signal. 2022. Vol. 39. P. 119–124.Jo B., Kim S. Comparative Analysis of OpenPose, PoseNet, and MoveNet Models for Pose Estimation in Mobile Devices // Trait. du Signal. 2022. Vol. 39. P. 119–124.Cao Z., Simon T., Wei S.E., Sheikh Y. Realtime multi-person 2d pose estimation using part affinity fields // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017. P. 7291–7299.Cao Z., Simon T., Wei S.E., Sheikh Y. Realtime multi-person 2d pose estimation using part affinity fields // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017. P. 7291–7299.Erramchetty S.K. et al. Human pose estimation using blaze pose // AIP Conference Proceedings. 2024. Vol. 2971. No. 1.Erramchetty S.K. et al. Human pose estimation using blaze pose // AIP Conference Proceedings. 2024. Vol. 2971. No. 1.Gadhiya R., Kalani N. Analysis of Deep Learning Based Pose Estimation Techniques for Locating Landmarks on Human Body Parts // Proceedings of the 2021 International Conference on Circuits, Controls and Communications (CCUBE). 2021. P. 1–4.Gadhiya R., Kalani N. Analysis of Deep Learning Based Pose Estimation Techniques for Locating Landmarks on Human Body Parts // Proceedings of the 2021 International Conference on Circuits, Controls and Communications (CCUBE). 2021. P. 1–4.Lin T.Y., Maire M., Belongie S., Hays J., Perona P., Ramanan D., Dollár P., Zitnick C.L. Microsoft coco: Common objects in context // Proceedings of the European Conference on Computer Vision. 2014. P. 740–755.Lin T.Y., Maire M., Belongie S., Hays J., Perona P., Ramanan D., Dollár P., Zitnick C.L. Microsoft coco: Common objects in context // Proceedings of the European Conference on Computer Vision. 2014. P. 740–755.Andriluka M., Pishchulin L., Gehler P., Schiele B. 2D human pose estimation: New benchmark and state of the art analysis // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014. P. 3686–3693.Andriluka M., Pishchulin L., Gehler P., Schiele B. 2D human pose estimation: New benchmark and state of the art analysis // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014. P. 3686–3693.Toshev A., Szegedy C. DeepPose: Human pose estimation via deep neural networks // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014. P. 1653–1660.Toshev A., Szegedy C. DeepPose: Human pose estimation via deep neural networks // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2014. P. 1653–1660.Chung J.-L., Ong L.-Y., Leow M.-C. Comparative Analysis of Skeleton-Based Human Pose Estimation // Future Internet. 2022. Vol. 14, No. 12. P. 380. https://doi.org/10.3390/fi14120380Chung J.-L., Ong L.-Y., Leow M.-C. Comparative Analysis of Skeleton-Based Human Pose Estimation // Future Internet. 2022. Vol. 14, No. 12. P. 380. https://doi.org/10.3390/fi14120380Ahmedt-Aristizabal D., Nguyen K., Denman S., Sridharan S., Dionisio S., Fookes C. Deep motion analysis for epileptic seizure classification // Proceedings of the 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 2018. P. 3578–3581.Ahmedt-Aristizabal D., Nguyen K., Denman S., Sridharan S., Dionisio S., Fookes C. Deep motion analysis for epileptic seizure classification // Proceedings of the 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. 2018. P. 3578–3581.Ishaq A., Haq I.U. Real-time single-person pose estimation on low-power embedded devices // 2019 International Conference on Frontiers of Information Technology (FIT).Ishaq A., Haq I.U. Real-time single-person pose estimation on low-power embedded devices // 2019 International Conference on Frontiers of Information Technology (FIT).Cao Z., Simon T., Wei S.-E., Sheikh Y. Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017. P. 7291–7299.Cao Z., Simon T., Wei S.-E., Sheikh Y. Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields // Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2017. P. 7291–7299.Papandreou G., Zhu T., Chen L.-C., Gidaris S., Tompson J., Murphy K. PersonLab: Person Pose Estimation and Instance Segmentation with a Bottom-Up, Part-Based, Geometric Embedding Model // Proceedings of the European Conference on Computer Vision (ECCV). 2018. P. 269–286.Papandreou G., Zhu T., Chen L.-C., Gidaris S., Tompson J., Murphy K. PersonLab: Person Pose Estimation and Instance Segmentation with a Bottom-Up, Part-Based, Geometric Embedding Model // Proceedings of the European Conference on Computer Vision (ECCV). 2018. P. 269–286.TensorFlow Blog. MoveNet: Ultra fast and accurate pose detection model. 2021. URL: https://blog.tensorflow.org/2021/05/movenet-and-posture-analysis.htmlTensorFlow Blog. MoveNet: Ultra fast and accurate pose detection model. 2021. URL: https://blog.tensorflow.org/2021/05/movenet-and-posture-analysis.htmlBazarevsky V., Zhang F. BlazePose: On-Device Real-Time Body Pose Tracking. Google AI Blog. 2020. URL: https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.htmlBazarevsky V., Zhang F. BlazePose: On-Device Real-Time Body Pose Tracking. Google AI Blog. 2020. URL: https://ai.googleblog.com/2020/08/on-device-real-time-body-pose-tracking.htmlGabdullina D.R., Zykov E.Yu., Kugurakova V.V. Using OpenCV for real-time image recognition through augmented reality devices // The 2023 International Conference on Artificial Life and Robotics (ICAROB2023). 2023. P. 868–874.Gabdullina D.R., Zykov E.Yu., Kugurakova V.V. Using OpenCV for real-time image recognition through augmented reality devices // The 2023 International Conference on Artificial Life and Robotics (ICAROB2023). 2023. P. 868–874.Gabdullina D.R., Zykov E.Yu., Kugurakova V.V. Using OpenCV for augmented reality appliсations // Journal of Advances in Artificial Life Robotics. 2023. Vol. 3. Iss. 3(3). P. 179–184.Gabdullina D.R., Zykov E.Yu., Kugurakova V.V. Using OpenCV for augmented reality appliсations // Journal of Advances in Artificial Life Robotics. 2023. Vol. 3. Iss. 3(3). P. 179–184.Бакиров А.Р. и др. Использование контроллера Microsoft Kinect в разработке реабилитационных игр // Электронные библиотеки. 2016. Т. 19. №. 6. С. 521–537.Бакиров А.Р. и др. Использование контроллера Microsoft Kinect в разработке реабилитационных игр // Электронные библиотеки. 2016. Т. 19. №. 6. С. 521–537.You Y., Zhao Y. A human pose estimation algorithm based on the integration of improved convolutional neural networks and multi-level graph structure constrained model // Pers. Ubiquitous Comput. 2019. Vol. 23. P. 607–616.You Y., Zhao Y. A human pose estimation algorithm based on the integration of improved convolutional neural networks and multi-level graph structure constrained model // Pers. Ubiquitous Comput. 2019. Vol. 23. P. 607–616.

The authors declare that there are no conflicts of interest present.