Object Detection and Tracking Using YOLO & SORT

Object detection and tracking are fundamental tasks in computer vision that enable real-time analysis of visual data in dynamic environments. This research paper presents a comprehensive study of a YOLOv11 and DeepSORT-based object detection and tracking system designed for robust performance in real-world scenarios. The proposed system integrates YOLOv11, one of the latest deep learning detection models, with DeepSORT, a sophisticated multi-object tracking algorithm that combines Kalman Filter motion prediction and appearance-based feature embeddings. Enhanced architectural components including C3k2 blocks, SPPF, and C2PSA attention modules ensure high accuracy detection even under challenging conditions such as occlusion, varying illumination, and rapid motion. The system achieves above 25 FPS on mid-range computing hardware while maintaining high detection accuracy and stable tracking. Applications range from surveillance and traffic monitoring to autonomous vehicles and intelligent security systems. This paper provides detailed architectural design, mathematical formulations, implementation methodology, and comprehensive experimental results demonstrating the effectiveness of the proposed approach.