AI-Based Personalized Fitness Trainer: AI Real-Time Pose Estimation and Form Correction Using Computer Vision

Suchetha N V; Skandan M J; Suchint M Shetty; Sagar D Patgar; Nikhil M C

doi:10.55524/ijircst.2025.13.6.11

Abstract

Over the past few years, workout tech was able to utilise smart software and camera tracking to enhance training alone. This paper presents a fitness assistant that is a browser-based implementation that works on AI, automatically counting the reps and identifying the incorrect postures in real-time using a standard web camera. It takes the place of special gear by using tools such as MediaPipe and TensorFlow to map significant body points when exercising. These systems accompany your limbs in examining the way in which joints bend and move as time goes on. When something is out of place, feedback appears immediately - assistance in correcting the technique in time. This correctly improves the accuracy of learning and reduces the risk of exercise injury. The device is paired with an interactive online training program to train, monitor progress or review previous outcomes - no wearable devices are required. It has passed tests with regards to its ability to detect correct form during the execution of such moves as squats, lunges, or push-ups and pose checks. The proposed option would be effective in terms of online fitness, monitoring progress in recovery, or upgrading gym equipment - it provides a low-cost solution to exercise progression smarter.

Keywords

Posture Analysis, Real-Time Feedback, Repetition Counting, Exercise Monitoring, Workout Tracking, MediaPipe Pose, Real-Time Detection, Pose Estimation, Skeleton Tracking, Motion Analysis

References

Chen, S., and R. R. Yang, “Pose trainer: correcting exercise posture using pose estimation,” arXiv preprint arXiv:2006.11718, 2020. Available from: https://doi.org/10.48550/arXiv.2006.11718
Riccio, R., “Real-Time Fitness Exercise Classification and Counting from Video Frames,” arXiv preprint arXiv:2411.11548, 2024. Available from: https://doi.org/10.48550/arXiv.2411.11548
Dsouza, D. Maurya, and A. Patel, “Smart gym trainer using human pose estimation,” in Proc. 2020 IEEE Int. Conf. for Innovation in Technology (INOCON), Bengaluru, India, 2020, pp. 1–4. Available from: https://ieeexplore.ieee.org/abstract/document/9298212
V. Bharath Kumar and A. Julian, “AI-Driven Fitness Coach: Webcam-based form correction and rep counting for optimized workouts,” in Proc. 2024 IEEE Int. Conf. on Computing, Power and Communication Technologies (IC2PCT), Greater Noida, India, 2024, pp. 399–402. Available from: https://ieeexplore.ieee.org/abstract/document/10486596
M. Chariar, S. Rao, A. Irani, S. Suresh, and C. S. Asha, “AI trainer: Autoencoder based approach for squat analysis and correction,” IEEE Access, vol. 11, pp. 107135–107149, 2023. Available from: https://ieeexplore.ieee.org/abstract/document/10254650
Y. Kwon and D. Kim, “Real-time workout posture correction using OpenCV and MediaPipe,” J. Korean Inst. Inf. Technol., vol. 20, no. 1, pp. 199–208, 2022. Available from: https://doi.org/10.14801/jkiit.2022.20.1.199
Y. Kwon and D. Kim, “Real-time workout posture correction using OpenCV and MediaPipe,” J. Korean Inst. Inf. Technol., vol. 20, no. 1, pp. 199–208, 2022. Available from: https://doi.org/10.14801/jkiit.2022.20.1.199
T. Shirsat, S. Melasagare, B. Charate, and S. Alte, “AI fitness trainer: Dynamic motion correction and real-time feedback for enhancing self-managed workouts,” pp. 1–6, 2024. Available from: https://ieeexplore.ieee.org/abstract/document/10581080
R. K. Dewang, A. Ali, D. Kumar, Ranvijay, A. Kumar, and S. Omar, “Real-time posture correction system in sitting using deep learning techniques,” in Proc. 2025 IEEE 14th Int. Conf. on Communication Systems and Network Technologies (CSNT), Bhopal, India, 2025, pp. 309–315. Available from: https://ieeexplore.ieee.org/abstract/document/10968856
L. Rothkrantz, “Personalized digital fitness coach,” in Proc. 22nd Int. Conf. on Computer Systems and Technologies, 2021. Available from: https://doi.org/10.1145/3472410.3472412
Kumar et al., “Human posture detection and correction using MediaPipe and OpenCV,” in Int. Conf. on Computing and Communication Systems for Industrial Applications, Singapore: Springer Nature, 2024. Available from: https://link.springer.com/chapter/10.1007/978-981-97-5862-3_11
S. Jin et al., “Whole-body human pose estimation in the wild,” in Proc. European Conf. on Computer Vision (ECCV), Cham: Springer, 2020. Available from: https://link.springer.com/chapter/10.1007/978-3-030-58545-7_12
Pardeshi et al., “Fitness freaks: A system for detecting definite body posture using OpenPose estimation,” in Futuristic Trends in Networks and Computing Technologies, Singapore: Springer Nature, 2022. Available from: https://link.springer.com/chapter/10.1007/978-981-19-5037-7_76
Nagarkoti et al., “Realtime indoor workout analysis using machine learning and computer vision,” in Proc. 41st Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. (EMBC), 2019. Available from: https://ieeexplore.ieee.org/abstract/document/8856547
Rohan et al., “Human pose estimation-based real-time gait analysis using convolutional neural network,” IEEE Access, vol. 8, pp. 191542–191550, 2020. Available from: https://ieeexplore.ieee.org/abstract/document/9220146
Samhitha et al., “Vyayam: Artificial intelligence-based bicep curl workout tracking system,” in Proc. 2021 Int. Conf. on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES), 2021. Available from: https://ieeexplore.ieee.org/abstract/document/9633841
V. G. Biradar Ajay et al., “AI tool as a fitness trainer using human pose estimation,” in Proc. 2023 Int. Conf. on Network, Multimedia and Information Technology (NMITCON), 2023. Available from: https://ieeexplore.ieee.org/abstract/document/10276320
S. B. Pauzi et al., “Movement estimation using MediaPipe BlazePose,” in Advances in Visual Informatics, IVIC 2021, Springer, 2021. Available from: https://tinyurl.com/yreexnxf
Y. Wang et al., “LitePose: Efficient architecture design for 2D human pose estimation,” in Proc. IEEE/CVF Conf. on Computer Vision and Pattern Recognition (CVPR), 2022. Available from: https://tinyurl.com/yc4v3fzm
Y. Liu, C. Qiu, and Z. Zhang, “Deep learning for 3D human pose estimation and mesh recovery: A survey,” Neurocomputing, vol. 596, 2024. Available from: https://doi.org/10.1016/j.neucom.2024.128049

Cites this article as

S. N. V, S. M. J, S. M. Shetty, S. D. Patgar, N. M. C, "AI-Based Personalized Fitness Trainer: AI Real-Time Pose Estimation and Form Correction Using Computer Vision", International Journal of Innovative Research in Computer Science and Technology (IJIRCST), Vol-13, Issue-6, Page No-101-110, 2025. Available from: https://doi.org/10.55524/ijircst.2025.13.6.11

Corresponding Author

Suchetha N V

Associate Professor, Department of Computer Science & Engineering, Sri Dharmasthala Manjunatheshwara Institute of Technology, Ujire, Karnataka, India

Download Full Paper

Download PDF

No. of Downloads: 4 | No. of Views: 31

A Comprehensive Review on Descriptive Answer Script Evaluation Techniques

Pradeep Rao K B, Thyagaraju G S, Sahana Kumari B, Prasad S R.

November 2025 - Vol 13, Issue 6
Explainable Artificial Intelligence (XAI) for Distributed Systems: A Survey

Rizwan Saleem, Nazia Akram, Misbah Aziz Rajput, Imran Mahmood.

November 2025 - Vol 13, Issue 6
Eduventure- A Gamified Microlearning Platform

Pradeep G S, Dhavalashree Jain L A, Pushpa B, Shubha Parameshwar Bhat, Soujanya.

November 2025 - Vol 13, Issue 6

AI-Based Personalized Fitness Trainer: AI Real-Time Pose Estimation and Form Correction Using Computer Vision

Citations

Download Full Paper PDF

Total View 31

Total Download 4