This project implements an eye gaze detection system using a particle filter and a neural network. The system captures video from a webcam, detects eyes using a Haar cascade classifier, and estimates the gaze direction by tracking particles around the detected eyes.
I did this project in MATLAB language, using particle filter and neural network techniques. The MATLAB code is located in the matlab folder. Additionally, a Python version of the project is provided in the python folder.
- Real-time eye detection using Haar cascades
- Particle filter-based gaze direction estimation
- Visualization of detected eyes and gaze direction
- Python 3.x
- OpenCV
- NumPy
- Matplotlib (optional, for visualization)
You can install the required Python libraries using pip:
!pip install opencv-python numpy matplotlib
-
Clone the repository:
git clone https://github.com/your-username/eye-gaze-detection.git
-
Navigate to the Python project directory:
cd eye-gaze-detection/python -
Run the Python script:
python eye_gaze_detection.py
-
View the video feed:
- The video feed will open in a window displaying detected eyes and the estimated gaze direction.
- Press 'q' to exit the video feed window.
matlab/: Contains the MATLAB code for the project.python/: Contains the Python code for the project, includingeye_gaze_detection.py.
NUM_PARTICLES: Number of particles used in the particle filter.PARTICLE_NOISE: Standard deviation of noise added to particle positions.PARTICLE_WEIGHT: Weight factor for gaze direction simulation.
For questions or feedback, please contact zsarayloo@gmail.com.
You can copy and paste this Markdown content into a README.md file for your project. This README includes a complete overview of your project, installation instructions, how to run the code, folder structure, particle filter parameters, and contact information.