Drone Eye-Tracking Project

Overview

This repository contains Python scripts for eye tracking and face detection, likely intended for drone navigation or similar applications. The project uses machine learning, computer vision libraries, and real-time video processing to achieve its goals.

Table of Contents

• Dependencies
• How to Run
  • Training the Eye Position Model
  • Real-Time Eye Position Prediction
  • Simple Eye and Face Detection
  • Eye Labeling
  • Alternative Eye and Face Detection
• Contributing
• License

Dependencies

• Python 3.x
• PyTorch
• OpenCV
• dlib
• PIL (Pillow)
• imutils

You can install these packages using pip:

```bash pip install torch torchvision opencv-python dlib Pillow imutils ```

How to Run

Training the Eye Position Model

Run trainer.py to train a Convolutional Neural Network (CNN) for eye position classification.

```bash python trainer.py ```

Real-Time Eye Position Prediction

Run main.py to perform real-time eye position prediction using the trained model.

```bash python main.py ```

Simple Eye and Face Detection

Run eye_tracking.py for a simpler version of eye and face detection using OpenCV's Haar Cascades.

```bash python eye_tracking.py ```

Eye Labeling

Run eye_labeler.py to label eye positions in real-time. This will save frames and labels for future use.

```bash python eye_labeler.py ```

Contributing

Feel free to fork this repository, create a new branch, and submit a pull request.

License

This project is licensed under the MIT License.