This project focuses on the development of a human-tracking algorithm that applies YOLOv8 for object detection combined with DeepSORT for tracking. The aim is to analyze and track the movement of individuals through a mall, using a short video dataset as the foundation for our detection and tracking model.
In crowded environments like malls, tracking the movement and behavior of individuals can be invaluable for security and retail analytics. This project leverages the YOLOv8 model for its robust object detection capabilities, alongside DeepSORT, an advanced tracking algorithm, to accurately follow individuals across video frames. Our objective is to demonstrate the efficacy of these technologies in real-world scenarios.
The dataset consists of short video clips capturing people moving through various areas of a mall.
Source: Kaggle - People Tracking
- Python
- YOLOv8 (You Only Look Once version 8)
- DeepSORT
- OpenCV (cv2)
- Object Detection: Implementing YOLOv8 for real-time object detection to identify individuals within the video frames.
- Object Tracking: Utilizing DeepSORT to track the detected individuals across the video sequence, ensuring continuity and accuracy in movement tracking.
- Video Processing: Applying OpenCV for handling video data, including reading, processing, and displaying video frames for analysis and tracking.
For more information or to discuss this project further, please reach out.
Email: [email protected]