📖 Overview
This project implements a robust real-time object detection and tracking pipeline using YOLOv11 and Deep SORT. It is designed to accurately detect and track multiple objects in live video streams or pre-recorded footage with persistent object IDs across frames. The core objective is to ensure consistent object identification, even during occlusions or camera movements, making it suitable for surveillance, vehicle tracking, and industrial monitoring.

🌟 Key Features
✅ Real-time Detection: Utilizes YOLOv11 for fast and accurate object detection with improved speed and precision over earlier YOLO versions.

🔄 Deep SORT Tracking: Employs the Deep SORT algorithm to assign unique IDs to objects and track them persistently across frames using appearance and motion cues.

🎯 Confidence Thresholding: Offers configurable detection confidence levels to balance between false positives and missed detections.

📐 Aspect Ratio Preserved: Maintains original aspect ratio during image resizing using letterbox padding to avoid distortion.

🎯 Central Tracking Filter: Optionally filters tracked objects to include only those within the central region of the frame (useful for focused analysis like crowd movement in a specific area).

🖼️ Live Video Support: Compatible with webcam or RTSP streams for live monitoring applications.

🔧 Installation & Usage
The repository includes all required setup instructions, YOLOv11 model loading, Deep SORT integration, and sample video processing. Just clone the repo, install dependencies, and run the main script to start detection and tracking.

🔗 Repository
GitHub: iamrukeshduwal/yolov11_real_time_object_detection_with_DeepSORT