🎮 Face-Controlled Game Controller using OpenCV

An AI-powered Python application that enables hands-free game control using real-time face detection via webcam.
Built with OpenCV DNN for detection and PyAutoGUI for keyboard automation, this project brings accessibility and interactivity to gaming and beyond.

---

✨ Features

• Real-time face detection using OpenCV DNN.
• Direction-based key control (⬅️ ⬆️ ➡️ ⬇️) mapped to head movement.
• FPS monitoring for performance tracking.
• Modular and extensible codebase.
• One-click game start when face enters control zone.

---

📂 Project Structure

face-game-controller/
│
├── assets/
│   ├── banner.png                                     # Project banner
│   └── sample_output.png            
│
├── models/                                            # Pretrained Caffe model
│   ├── deploy.prototxt
│   └── res10_300x300_ssd_iter_140000.caffemodel
│ 
├── utils/
│   ├── config.py                                      # Configuration (model paths) 
│   ├── controller.py                                  # Movement control (PyAutoGUI) 
│   └── face_detection.py                              # Face detection logic (OpenCV DNN)
│ 
├── main.py
│ 
├── LICENSE
├── requirements.txt                                   # Dependencies
├── .gitignore
└── README.md

---

⚙️ Installation

# Clone repository
git clone https://github.com/HeleenaRobert/face-game-controller-opencv.git
cd face-game-controller-opencv

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

# Install dependencies
pip install -r requirements.txt

---

📦 Model Files

The project includes the OpenCV DNN Face Detector based on a Caffe model.
These files are already provided in the models/ folder:

• deploy.prototxt
• res10_300x300_ssd_iter_140000.caffemodel

No additional downloads are required.

---

🚀 Usage

python main.py

• Align your face in the control zone to start.
• Move your head left / right / up / down → triggers arrow key events.
• Press ESC to exit.

---

🎯 How It Works

1. Loads OpenCV DNN face detection model.
2. Captures webcam frames and detects face positions.
3. Defines a central control zone.
4. Detects movement → triggers directional key events using PyAutoGUI.
5. Logs activity and displays FPS overlay.

---

📊 Sample Output

When running the program, the webcam feed will open with:

• A green bounding box highlighting the detected face.
• A red central control zone where the face must align to start the game.
• FPS counter shown on top-left.
• Head movements mapped to key events:

Face Position	Triggered Action
Face inside control zone	Game starts (auto click)
Move head left ⬅️	Presses Left Arrow
Move head right ➡️	Presses Right Arrow
Move head up ⬆️	Presses Up Arrow
Move head down ⬇️	Presses Down Arrow

Example annotated frame:

---

🛠️ Future Improvements

• GUI frontend (PyQt5).
• Multi-user support.
• Game scoring/session logging.
• REST API integration.

---

📝 License

This project is licensed under the MIT License.

---

👩‍💻 Author

Heleena Robert
GitHub