This project is a Python PyQt5-based graphical interface that: ✅ Selects between IP, RTSP, or USB cameras
✅ Runs YOLOv8 object detection (supports small, medium, large models)
✅ Displays human, car, and ship detections with count icons
✅ Supports GPU (CUDA) or CPU inference
✅ Saves user settings and last-used camera configuration
✅ Includes a splash screen with configurable app logo

• ✅ Real-time video stream from USB, IP, or RTSP cameras
• ✅ Model switcher (YOLOv8s, YOLOv8m, YOLOv8l) for performance vs. accuracy
• ✅ Sidebar with toggleable object detections: human, car, ship
• ✅ Right-click on video window: toggle detection lines, change resolution (HD, FHD, 4K)
• ✅ Menu bar: switch processing mode (CPU/GPU), change logo, save settings
• ✅ Configuration saved in config.json for session persistence

# Clone the repo
git clone https://github.com/m7mds91/YOLO-Detection.git
cd YOLO-Detection

# Create virtual environment (optional but recommended)
python -m venv venv
source venv/bin/activate  # On Windows: venv\\Scripts\\activate

# Install required packages
pip install -r requirements.txt

🛠 Example requirements.txt

PyQt5==5.15.9
opencv-python>=4.8.0
ultralytics>=8.0.0
torch>=2.0.0

🚀 Run the app

python YOLO-Detection.py

⚙️ Configuration

Logo icon: Set/change via the menu (Settings → Change Logo)

Model:

Switch via menu (Performance → YOLOv8s / YOLOv8m / YOLOv8l)
Processing device: CPU or GPU (if available)
Saved settings: Stored in config.json in the project folder

📸 Supported Cameras

💻 Notes

For GPU support, ensure PyTorch with CUDA is installed matching your system and GPU drivers.
Check: PyTorch Install Guide
YOLOv8 models (.pt files) should be placed in the project directory or adjust the script paths.

🌟 Credits

YOLOv8: Ultralytics PyQt5 GUI: Riverbank Computing