Face Mask AR is a Python program that detects faces in a live video stream and augments them with a mask image. This provides a fun and interactive way to try on different masks virtually.
- Python 3
- OpenCV library (
pip install opencv-python) - Haar cascade XML file for face detection (e.g.,
haarcascade_frontalface_default.xml) taken fromhttps://github.com/kipr/opencv/blob/master/data/haarcascades/haarcascade_frontalface_default.xml - Mask image with transparency (RGBA format)
-
Clone the repository or download the Python script and the required image files.
-
Install the necessary libraries by running the following command in the terminal:
pip install opencv-python -
Place the Haar cascade XML file (
haarcascade_frontalface_default.xml) and the mask image (e.g.,mask.png) in the same directory as the Python script. -
Run the Python script in a terminal or IDE:
python face_mask_ar.py -
The program will open a window showing the live video stream from your default webcam. It will detect faces in the video and augment them with the selected mask image.
-
To choose a different mask, press the "Q" key to exit the program and update the
mask.pngfile with your desired mask image.
Example screenshot of the program displaying the live video stream with a detected face and a mask augmented on it.
- Ensure that your webcam is properly connected and accessible by the program.
- Experiment with different mask images to achieve the desired augmented reality effect.
- Adjust the face detection parameters in the script (e.g., scale factor, minimum neighbors) for better face detection results if needed.