This is a simple program that detects faces in a video feed using the Dlib library and its pre-trained models. The detected faces are then drawn on the frame using their facial landmarks.
To run this program, you will need:
- Python 3.x
- OpenCV
- Dlib
- NumPy
Clone the repository and install the required libraries using pip:
https://github.com/wajdisawa/face_detection
Make sure you have the pre-trained models downloaded and saved in the same directory as the script. The models are:
shape_predictor_81_face_landmarks.datshape_predictor_68_face_landmarks.dat
You can download these models from the Dlib website.
To run the program, execute the following command:
python detect_faces.py
This will open a window displaying the video feed with the detected faces and their corresponding facial landmarks. The program will terminate if the 'q' key is pressed.
This project is licensed under the MIT License - see the LICENSE file for details.