Originally, I followed the face detection guide for Python provided by Mediapipe, where I created a face detector instance. Face detection was a success, yet I could not find any ways to obtain the variables origin_x, origin_y, height and width from the FaceDetectorResult object so as to blur the face. Thus, I resorted to using mp.solutions.face_detection with easy-to-use APIs to process frames and obtian face detection results, such as the coordinates of the bounding boxes around the detected faces.
Simply run the program (after installing the libraries - OpenCV and Mediapipe) and if you're in front of the camera, your face will be blurred. To stop the program from running, press x on your keyboard.
The idea was inspired by Murtaza Hassan's video, yet the code is different.