Requirements:
Must have Dlib shape_predictor_68_face_landmarks.dat.bz2
See requirments.txt for packages
once installed run detection.py
Proper use: Create account Set face in normal gaming position including eyes focused on middle/ bottom 2/3 of screen. Change sensitivity, the higher the number out of 10 the more difficult it triggers that you are on your phone Capture face with S(only stored locally) Game like normal and visit (https://phoneleaderboard.andrewslayton.dev/) to see how often you are on your phone compared to others while gaming/using your computer
How phoneChecker works: phone checker uses dlib in combination with the dataset mentioned above to predict the users pupils, forehead, and chin. I run a simple calculation determined by thousands of instances of data I acquired by testing on my own and multiple volunteer testers. if you are determined to be on your phone for 30 or more seconds, your phone tally count is incremented by one. The data is then sent to the website to be displayed
DISCLAIMER: Facial data is not saved, this is honor system based please do not track under non created usernames.