SmartShuttle: AI-Based Badminton Player Performance Analysis Using Computer Vision

In order to overcome the drawbacks of conventional manual approaches, which are laborious, subjective, and prone to error. In order to accomplish this, we created a comprehensive deep learning system that combines CNNs for shot classification, OpenPose for pose prediction, YOLOv5 for racket detection, and Hough Transform for court line extraction. Our system achieved 92.5% accuracy for shot classification, 86.5% dependability for pose prediction, and 93.7% precision for court recognition. It was also scalable in terms of the valuable insights it offered (biomechanical metrics; tactical heatmaps).  However, the approach has a number of drawbacks, including real-time processing restrictions, occlusions during player-net interaction, and the dataset is only accessible to professional players (18fps). Therefore, future approaches should concentrate on refining the framework for real-time deployment on edge devices, expanding the data set to cover a wider range of skill levels and play scenarios, and enhancing occlusion management with more intricate designs like transformers. In the future, the system might be expanded to track player interactions and weapon ballistics using wearable sensors. These advancements would strengthen the framework's ability to support sports analytics as a more objective, scalable, and coach-friendly tool.

[1]   R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction. Cambridge, MA, USA: MIT Press, 2018.  [2]        D. Silver, A. Huang, C. J. Maddison, A. Guez,L. Sifre, and G. Van Den Driessche, “Mastering the game of Go with deep neural networks and tree search,” Nature, vol. 529, no. 7587, pp. 484– 489,2016.  [3]     A. E. Sallab, M. Abdou, E. Perot, and S. Yogamani, “Deep reinforcement learning framework for autonomous driving,” Electron. Imag., vol. 2017, no. 19, pp. 70–76, 2017.   [4]   Y. Liu, B. Logan, N. Liu, Z. Xu, J. Tang, andY. Wang, “Deep reinforcement learning for dynamic treatment regimes on medical registry data,” in Proc. IEEE Int. Conf. Healthcare Informat. (ICHI), Aug. 2017, pp. 380–385.  [5]     S. Gu, E. Holly, T. Lillicrap, and S. Levine, “Deep reinforcement learning for robotic manipulation with asynchronous off-policy updates,” in Proc. IEEE Int. Conf. Robot.Autom. (ICRA), May 2017, pp. 3389–3396  [6]       M. Hausknecht and P. Stone, “Deep recurrent Q-learning for partially observable MDPs,” in Proc. AAAI Fall Symp. Ser., 2015,pp. 1–9. [7]   V. Mnih, K. Kavukcuoglu, D. Silver, A. A. Rusu, J. Veness, M. G. Bellemare, and A. Graves, “Human-level control through deep reinforcement learning,” Nature, vol. 518, no. 7540, pp. 529–533, 2015.  [7]    R. S. Sutton and A. G. Barto, Reinforcement Learning: An Introduction. Cambridge, MA, USA: MIT Press, 2018.  [8]        D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, and G. Van Den Driessche, “Mastering the game of Go with deep neural networks and tree search,” Nature, vol. 529, no. 7587, pp. 484– 489, 2016.  [9]     A. E. Sallab, M. Abdou, E. Perot, and S. Yogamani, “Deep reinforcement learning framework for autonomous driving,” Electron. Imag., vol. 2017, no. 19, pp. 70–76,2017.   [10]                 Y. Liu, B. Logan, N. Liu, Z. Xu, J. Tang, and Y. Wang, “Deep reinforcement learning for dynamic treatment regimes on medical registry data,” in Proc. IEEE Int. Conf. Healthcare Informat. (ICHI), Aug. 2017, pp. 380–385.