AIU – Faculty of Informatics and Communications Engineering
Published May 4, 2026 | Version v1
Preprint Open

GymFlow Pro: Reimagining Gym Management Through Computer Vision, Reinforcement Learning, and Intelligent Coaching

Authors/Creators

  • 1. ROR icon Arab International University

Description

GymFlow Pro is a comprehensive AI-powered mobile application designed to revolutionize the fitness experience for gym members, coaches, and administrators. Built using Flutter, Django, and PostgreSQL, the platform integrates four advanced AI technologies — Computer Vision for real-time exercise form correction, Reinforcement Learning for adaptive workout scheduling, Random Forest ensemble methods for body composition analysis, and Collaborative Filtering combined with NLP for personalized meal planning. The system addresses the critical shortcomings of existing fitness platforms, which rely on static content, manual data entry, and generic recommendations that fail to adapt to individual user needs. Beyond AI-driven training and nutrition guidance, GymFlow Pro offers an integrated gym management ecosystem including class booking, coach-client interaction, and a gamification system featuring leaderboards and achievement badges to sustain long-term engagement. The platform's AI models achieve strong projected accuracies of 89% in exercise form detection, 87% in workout scheduling, 85% in body composition analysis, and 88% in meal plan recommendation. GymFlow Pro represents a technically advanced and practically complete solution that unifies intelligent fitness coaching, personalized nutrition, and gym operations within a single seamless mobile platform. 

This work was conducted at Arab International University (AIU), Syria. The official website of the university is: https://www.aiu.edu.sy

Files

Basheer_2026_GymFlow Pro.pdf

Files (1.5 MB)

Name Size
md5:a2c515e95fc6d0ba18774d47ab8fdd69
1.5 MB Preview Download

Additional details

References

  • [1] T. Barhoum, "SETF: A Structured Engineering Thesis Framework for Artificial Intelligence, Software Engineering, and Robotics," Zenodo, Apr. 2026. doi: 10.5281/zenodo.19686845.
  • [2] J. Kim, S. Park, and H. Lee, "Real-Time Exercise Form Analysis Using Computer Vision and Pose Estimation for Mobile Fitness Applications," IEEE Transactions on Human-Machine Systems, vol. 54, no. 2, pp. 312–324, 2024.
  • [3] C. Ramirez, M. Torres, and A. Garcia, "Personalized Workout Scheduling Using Reinforcement Learning in Mobile Fitness Platforms," Journal of Ambient Intelligence and Smart Environments, vol. 15, no. 4, pp. 389–403, 2023.
  • [4] P. Hernandez, L. Moreno, and R. Castro, "Non-Invasive Body Composition Estimation Using Random Forest Ensemble Learning and Anthropometric Measurements," Computers in Biology and Medicine, vol. 162, Art. no. 107089, 2023.
  • [5] K. Tanaka, Y. Suzuki, and M. Yamamoto, "Collaborative Filtering for Personalized Dietary Recommendation in Mobile Health Applications," ACM Transactions on Information Systems, vol. 42, no. 1, pp. 1–28, 2024.
  • [6] E. Okafor, B. Adeyemi, and F. Nwosu, "Automated Dietary Preference Extraction from Unstructured Text Using Transformer-Based NLP Models," Expert Systems with Applications, vol. 238, Art. no. 122045, 2023.
  • [7] D. Morales, A. Jimenez, and V. Ruiz, "Impact of Gamification on Long-Term User Engagement and Fitness Behavior Change in Mobile Fitness Applications," Computers in Human Behavior, vol. 148, Art. no. 107912, 2024.
  • [8] R. Fernandez, S. Lopez, and J. Martinez, "Integrated Gym Management System Design for Enhanced Member Retention and Operational Efficiency," International Journal of Sports Management and Marketing, vol. 23, no. 3, pp. 201–218, 2023.
  • [9] T. Nakamura, H. Watanabe, and K. Ito, "Performance Evaluation of Flutter for AI-Powered Mobile Health Applications with Real-Time Processing Requirements," Mobile Networks and Applications, vol. 29, no. 2, pp. 445–461, 2024.