Face Anonymizer with OpenCV and MediaPipe

This project demonstrates how to detect faces in real-time from a webcam feed and blur them using OpenCV and MediaPipe. It leverages the MediaPipe Face Detection model to identify faces in a frame, and then applies a blur effect to those areas. Optionally, it can also draw bounding boxes around detected faces.

Requirements

To run the project, you'll need to install the following Python libraries:

• mediapipe: For face detection.
• opencv-python: For image processing and webcam access.

You can install the dependencies using pip:

pip install mediapipe opencv-python

Alternatively, you can install all required dependencies from the requirements.txt file by running:

pip install -r requirements.txt

Functions

blur_faces(frame, ksize: int, draw_bbox: bool=False)

This function detects faces in a given frame and applies a blur effect on the face regions.

Parameters:

• frame (numpy.ndarray): The image frame from the webcam or video source.
• ksize (int): The size of the kernel used for the blur effect.
• draw_bbox (bool, optional): If set to True, it will draw a bounding box around the detected face. Defaults to False.

Returns:

• A frame with blurred faces (and optionally bounding boxes).

main()

This function captures video from the default webcam, detects faces, applies the blur effect, and displays the resulting video with blurred faces in real-time. The video stream continues until the user presses the 'q' key to quit.

How It Works

1. Face Detection: MediaPipe's FaceDetection model is used to detect faces in each frame from the webcam.
2. Blurring Faces: Once a face is detected, a bounding box is computed based on the relative coordinates. The region inside the bounding box is then blurred using OpenCV's cv2.blur() function.
3. Display: The modified frame is displayed with the blurred faces, and optionally, bounding boxes can be drawn around the faces.

Usage

To run the program, simply execute the script. It will open the webcam feed and apply the face blurring effect in real-time.

python face_anonymizer.py

Press 'q' to close the webcam feed and exit the program.