Authors :
R. C. Mahajan; Bhupesh Cholake; Chinmayee More; Aryan Varkhede; Anushka Shirsath
Volume/Issue :
Volume 11 - 2026, Issue 5 - May
Google Scholar :
https://tinyurl.com/ky2earym
Scribd :
https://tinyurl.com/yxb358xx
DOI :
https://doi.org/10.38124/ijisrt/26may2242
Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.
Abstract :
This project proposes a practical, web-based intelligent coaching framework that integrates browser-side human
pose estimation, rule-based biomechanical reasoning, natural-language coaching through large language models, adaptive
workout generation, and nutrition planning into a unified digital health platform. The system is designed to provide live
exercise assessment and dance guidance without training custom deep posture-classification models. Instead, it leverages
pretrained pose landmark detectors and a layered decision architecture consisting of pose extraction, movement
interpretation, safety filtering, scoring, and personalized coaching. The project sits at the intersection of human-computer
interaction, computer vision, intelligent tutoring systems, digital therapeutics, sports analytics, and personalized health
informatics.
Keywords :
AI Fitness Coach; Human Pose Estimation; Posture Correction; Dance Analysis; Explainable AI.
References :
- International Journal of Creative Research Thoughts, “AI-Based Posture Correction Systems,” IJCRT, vol. 13, no. 1, pp. 1–6, 2025.
- IOSR Journal of Computer Engineering, “Computer Vision for Human Pose Estimation,” IOSR-JCE, vol. 27, no. 2, pp. 15–22, 2025.
- Y.-M. Ko, A. Nasridinov, and S.-H. Park, “Real-Time AI Posture Correction for Powerlifting Exercises Using YOLOv5 and MediaPipe,” IEEE Access, vol. 12, pp. 195830–195842, 2024.
- L. Xiang and G. Gao, “Perceptual Feature Integration for Sports Dancing Action Scenery Detection and Optimization,” IEEE Access, vol. 11, pp. 112345–112356, 2023.
- R. Nair, A. Yang, and H. Zhou, “AI-Powered Dance Coaching via Pose Estimation, Vision Transformers, and DTW,” in Proc. IEEE Int. Conf. on AI and Computer Vision, 2025, pp. 210–218.
- S. Gurbuxani, L. Gupta, and K. Madan, “Virtual Fitness Trainer Using Artificial Intelligence,” Int. J. Eng. Res. Technol. (IJERT), vol. 13, no. 2, pp. 45–50, 2024.
- “Machine Learning for Real-Time Exercise Correction and Injury Prevention: A Systematic Review,” Sensors, vol. 24, no. 3, pp. 1120–1135, 2024.
This project proposes a practical, web-based intelligent coaching framework that integrates browser-side human
pose estimation, rule-based biomechanical reasoning, natural-language coaching through large language models, adaptive
workout generation, and nutrition planning into a unified digital health platform. The system is designed to provide live
exercise assessment and dance guidance without training custom deep posture-classification models. Instead, it leverages
pretrained pose landmark detectors and a layered decision architecture consisting of pose extraction, movement
interpretation, safety filtering, scoring, and personalized coaching. The project sits at the intersection of human-computer
interaction, computer vision, intelligent tutoring systems, digital therapeutics, sports analytics, and personalized health
informatics.
Keywords :
AI Fitness Coach; Human Pose Estimation; Posture Correction; Dance Analysis; Explainable AI.