Intelligent Athletic Performance Monitoring System for Player Movement Detection and Step Analysis Using Yolo

Sports analytics has revolutionized the understanding of player performance, game strategies, and fan engagement. Despite its benefits, implementing real-time sports analytics remains challenging due to the high computational demands and extensive setup requirements of existing models. The limitations of these models present serious obstacles to their implementation on resource-constrained devices, including edge or mobile systems. Using the YOLO (You Only Look Once) object detection framework, Which realize Precision, Recall, The work focuses on detecting and bounding football players the ball in real-time match scenarios. A meticulously annotated dataset was created, and implementation was carried out. When the IoU threshold was set at 0.5 (mAP 0.5), the proposed system achieved a mean Average Precision (mAP) of 60.7%; when the IoU threshold was set at 0.95 (mAP 0.5:0.95), the mAP was 38.3%. The results support the model's capability of accurately detecting and tracking objects in real time. This research has shown that the system is effective in solving the standard problem in sports analytics through an easy option in tracking players and balls on devices with limited resources.