Authors : Rajashekar K. J.; Hanumanthappa S.; Chandan S. L.; Ranjitha H.; Ruchitha T. R.; Sinchana S.

Volume/Issue : Volume 10 - 2025, Issue 12 - December

Google Scholar : https://tinyurl.com/pxvckr3f

Scribd : https://tinyurl.com/5j84xz79

DOI : https://doi.org/10.38124/ijisrt/25dec068

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 30 to 40 days to display the article.

Abstract : Real-time exercise analysis can be successfully supported by MediaPipe BlazePose, spatiotemporal models, and machine-learning-driven form classification, according to recent research on pose-estimation-based fitness systems. Existing research shows that lightweight models have potential for posture correction and repetition counting, but it also highlights issues with accuracy, generalization, and user-specific adaptability. In this paper, we offer a MediaPipe-based workout monitoring system that employs the BlazePose model to calculate joint angles, extract 33 body landmarks, and assess exercise form using machine-learning classification and heuristic thresholds. Through an interactive interface, the suggested system offers real-time rep counting, posture feedback, and calorie prediction. With this method, customers can work out precisely and safely without the need for a personal trainer.

Keywords : KNN, OpenCV, MediaPipoe, Blazepose.

References :

1. Ching-Hang Chen and Dev Ramanan 3D Pose Estimation = 2D Pose Estimation + Matching In CVPR,2017
2. Martinez, Juliet, et al. "A simple yet effective baseline for 3d human pose estimation." Proceedings of the IEEE International Conference on Computer Vision. 2017.
3. Qingtian Yu, Haopeng Wang, Fedwa Laamarti and Abdulmotaleb El Saddik, A. Deep Learning-Enabled Multi Task System for Exercise Recognition and Counting. Multimodal Technol. Interact. 2021, 5, 55. https://doi.org/10.3390/mti5090055
4. Choi, Sangbum, Seokeon Choi, and Changick Kim. "MobileHumanPose: Toward Real-Time 3D Human Pose Estimation in Mobile Devices." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021. (Choi, 2021)
5. Xie, Ling, and Xiao Guo. "Object Detection and Analysis of Human Body Postures Based on TensorFlow." 2019 IEEE International Conference on Smart Internet of Things (SmartIoT). IEEE, 2019. (Xie, 2019)
6. G ̈ul, Varol1, Duygu Ceylan2 Bryan Russell2 Jimei Yang BodyNet: Volumetric Inference of 3D Human Body Shapes.
7. Zhe cao, Gines Hidalgo, Tomas simon, Shih-En Wei, Yaser Sheikh. OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields. Cs.CV, 2018.
8. Amit Nagarkoti, Revant Teotia, Amith K. Mahale and Pankaj K. Das. Real time Indoor Workout Analysis Using Machine Learning Computer Vision [2022].