Programming for Artificial Intelligence | Developed by ABDUL SALAM
Vision-X Tracker is a high-performance computer vision application that combines YOLOv8 for object detection and Deep SORT for real-time tracking. Featuring a "Brilliant Look" Cyber-Tech UI, this tool is designed for security monitoring, traffic analysis, and general object intelligence.
- Triple-Stream Intelligence: Supports Real-time Webcam, Video File Uploads, and Static Image Analysis.
- Advanced Tracking: Assigns unique Tracking IDs to objects (e.g., Person #1, Car #23) that persist across frames.
- Cyber-Tech Dashboard: A custom CSS-styled interface with glowing bounding boxes and live metric cards.
- Class Filtering: Multi-select options to focus only on specific objects like "Person," "Car," or "Motorcycle".
- Dynamic Analytics: Real-time count of total objects and specific category totals displayed in a glassmorphism dashboard.
- Model: YOLOv8 (Ultralytics) for state-of-the-art detection.
- Tracking: Deep SORT (Simple Online and Realtime Tracking) for ID persistence.
- Frontend: Streamlit for a responsive, modern web interface.
- Processing: OpenCV (Headless) and PyTorch.
Check out the live application here:
abdul-salam-vision.streamlit.app
To run this project on your local machine, follow these steps:
- Clone the repository:
git clone https://github.com/salamlakhan7/VisionX-Tracker.git
cd VisionX-Tracker
- Install dependencies:
pip install -r requirements.txt
- Run the application:
streamlit run app.py
app.py: The main application script containing detection logic and UI styling.requirements.txt: List of Python dependencies for cloud deployment.packages.txt: System-level dependencies for Linux servers.yolov8n.pt: Pre-trained YOLOv8 Nano model weights.
Project Portfolio | 2026