. Human Pose Estimation (HPE) is one of the important areas of computer vision, which serves to identify the joints of the human body based on image and video data. This technology is of great importance in the field of sports for analyzing athletes' movements, assessing exercise techniques, and preventing injuries. In this study, the influence of input image size, lighting conditions, and model architecture on the accuracy of joint point detection in the process of analyzing sports movements was studied. The results obtained will serve to improve the efficiency of human pose detection systems and expand the possibilities of their application in sports analysis. The performance of the YOLOv8n-pose, HRNet, and OpenPose models was evaluated under different image sizes (256×256, 384×384, and 640×640) and lighting conditions (normal, low, and high), and the analysis was performed based on the PCK (Percentage of Correct Keypoints), OKS (Object Keypoint Similarity), AUC (Area Under Curve), and NME (Normalized Mean Error) metrics. The results showed that increasing the image size leads to an improvement in the accuracy of joint point detection. At 640×640 resolution, the accuracy increased and the NME values decreased. The lighting conditions also had a significant impact on the model performance, with the most stable results observed under normal lighting conditions. The experimental results showed that the HRNet model demonstrated high accuracy results, the OpenPose model performed stably, and the YOLOv8n-pose model was an effective solution for real-time systems. The results confirm the importance of choosing optimal image parameters in HPE systems.
[1]. Zhang, Z.; Shin, S.-Y. Two-Dimensional Human Pose Estimation with Deep Learning: A Review. Appl. Sci. 2025, 15, 7344. https://doi.org/10.3390/app15137344
[2]. Stenum, J.; Cherry-Allen, K.M.; Pyles, C.O.; Reetzke, R.D.; Vignos, M.F.; Roemmich, R.T. Applications of Pose Estimation in Human Health and Performance across the Lifespan. Sensors 2021, 21, 7315. https://doi.org/10.3390/s21217315
[3]. Gong, W.; Zhang, X.; Gonzàlez, J.; Sobral, A.; Bouwmans, T.; Tu, C.; Zahzah, E.-h. Human Pose Estimation from Monocular Images: A Comprehensive Survey. Sensors 2016, 16, 1966. https://doi.org/10.3390/s16121966
[4]. Chung, J.-L.; Ong, L.-Y.; Leow, M.-C. Comparative Analysis of Skeleton-Based Human Pose Estimation. Future Internet 2022, 14, 380. https://doi.org/10.3390/fi14120380
[5]. Bulat, Adrian & Tzimiropoulos, Georgios. (2016). Human Pose Estimation via Convolutional Part Heatmap Regression. 9911. 10.1007/978-3-319-46478-7_44.
[6]. Newell, A., Yang, K., Deng, J. (2016). Stacked Hourglass Networks for Human Pose Estimation. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds) Computer Vision – ECCV 2016. ECCV 2016. Lecture Notes in Computer Science(), vol 9912. Springer, Cham. https://doi.org/10.1007/978-3-319-46484-8 29
[7]. Samkari, E.; Arif, M.; Alghamdi, M.; Al Ghamdi, M.A. Human Pose Estimation Using Deep Learning: A Systematic Literature Review. Mach. Learn. Knowl. Extr. 2023, 5, 1612-1659. https://doi.org/10.3390/make5040081
[8]. Sun & Xiao, Bin & Liu, Dong & Wang, Jingdong. (2019). Deep High-Resolution Representation Learning for Human Pose Estimation. 5686-5696. 10.1109/CVPR.2019.00584.
[9]. Cao, Zhe & Simon, Tomas & Wei, Shih-En & Sheikh, Yaser. (2017). Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields. 1302-1310. 10.1109/CVPR.2017.143.
[10]. Mundt, M.; Born, Z.; Goldacre, M.; Alderson, J. Estimating Ground Reaction Forces from Two-Dimensional Pose Data: A Biomechanics-Based Comparison of AlphaPose, BlazePose, and OpenPose. Sensors 2023, 23, 78. https://doi.org/10.3390/s23010078
[11]. Lu, T.; Cheng, K.; Hua, X.; Qin, S. KSL-POSE: A Real-Time 2D Human Pose Estimation Method Based on Modified YOLOv8-Pose Framework. Sensors 2024, 24, 6249. https://doi.org/10.3390/s24196249
[12]. Jeon, S.-J.; Park, S.Y.; Kim, S.T. Temperature Compensation of Fiber Bragg Grating Sensors in Smart Strand. Sensors 2022, 22, 3282. https://doi.org/10.3390/s22093282
[13]. Salem, M.A.; Sobaih, A.E.E. A Quadruple “E” Approach for Effective Cyber-Hygiene Behaviour and Attitude toward Online Learning among Higher-Education Students in Saudi Arabia amid COVID-19 Pandemic. Electronics 2023, 12, 2268. https://doi.org/10.3390/electronics12102268
[14]. Helmstetter, S.; Matthiesen, S. Human Posture Estimation: A Systematic Review on Force-Based Methods – Analyzing the Differences in Required Expertise and Result Benefits for Their Utilization. Sensors 2023, 23, 8997. https://doi.org/10.3390/s23218997
[15]. Gu, Y.; Zhang, H.; Kamijo, S. Multi-Person Pose Estimation using an Orientation and Occlusion Aware Deep Learning Network. Sensors 2020, 20, 1593. https://doi.org/10.3390/s20061593
[16]. Sun, R.; Lin, Z.; Leng, S.; Wang, A.; Zhao, L. An In-Depth Analysis of 2D and 3D Pose Estimation Techniques in Deep Learning: Methodologies and Advances. Electronics 2025, 14, 1307. https://doi.org/10.3390/electronics14071307
[17]. Ali, A.; Samara, W.; Alhaddad, D.; Ware, A.; Saraereh, O.A. Human Activity and Motion Pattern Recognition within Indoor Environment Using Convolutional Neural Networks Clustering and Naive Bayes Classification Algorithms. Sensors 2022, 22, 1016. https://doi.org/10.3390/s22031016
[18]. Duangburong, S.; Phruksaphanrat, B.; Muengtaweepongsa, S. Comparison of ANN and ANFIS Models for AF Diagnosis Using RR Irregularities. Appl. Sci. 2023, 13, 1712. https://doi.org/10.3390/app13031712
[19]. Gochoo, M.; Alnajjar, F.; Tan, T.-H.; Khalid, S. Towards Privacy-Preserved Aging in Place: A Systematic Review. Sensors 2021, 21, 3082. https://doi.org/10.3390/s21093082
[20]. Hammes, N.; Pinheiro, C.; Segundo, I.R.; Homem, N.C.; Silva, M.M.; Felgueiras, H.P.; Soares, G.M.B.; Freitas, E.; Costa, M.F.M.; Carneiro, J.A.O. Coaxial Fibres Incorporated with Phase Change Materials for Thermoregulation Applications. Appl. Sci. 2024, 14, 2473. https://doi.org/10.3390/app14062473