Object Detection, Tracking, and Prediction Analysis System for Sports Performance and Coaching
Swati Sunil Jagtap, Nilkanth B. Chopade, Viraj Deshmukh, Kasturi Dhavale, Pravin Hange · 2023
Object detection and tracking systems have diverse real-world applications, including surveillance and security, autonomous vehicles, robotics, sports analysis, and medical imaging. These systems can detect and track objects of interest, providing valuable insights into various domains and enabling autonomous operation. Current object analysis systems often have limitations in accuracy and, efficiency, and realtime processing capabilities. These limitations include occlusion, motion blur, lighting conditions, scale and orientation, computational complexity, false positives and false negatives, and limited object types. These limitations can affect the applicability of object analysis systems in various domains and limit their ability to detect and track objects accurately. The proposed system aims to overcome these limitations by leveraging the powerful processing capabilities of Jetson Xavier and combining the Kalman filter and polynomial regression techniques for more accurate and efficient object analysis. The proposed system uses state-of-the-art computer vision techniques for object tracking and trajectory prediction. The proposed approach can be valuable in sports analysis, providing accurate and efficient object detection, tracking, and prediction analysis.