AI Screen Time Measurement System
AI system to track your and your family's screen time more accurately.
Be Healthy, Be Cool!
Track your and your family's time more accurately, take one step towards being healthy
AI Screen Time will enable you to measure the time in which you were looking towards the screen, it will not measure the screen time when you are away from the screen, thus measuring your screen time More accurately.
V I D E O
ai-mentorship-camp.mp4The Context of the project
In this period of Pandemic, the screen time of everyone has been increased, controlling the screen time is very important, to control the time we should first monitor it, we have apps for measuring the screen time but there is a problem, many devices are shared by family members, the screen time app just measures the total time of the screen used but not for the individual member.
So to find the solution for measuring the screen time of each member using the common device, we came up with an idea about making an AI-enabled app for measuring screen time, which will be able to measure the screen time of an individual member using the device. This application uses Computer-Vision (CV) for its work.
v1 - Haar Cascades
The first version of The Project Used custom Haar Cascades.
v2 - Dlib
With dlib it's now more accurate, less resource consumption, and less data needed for proper working.
How It works?
AI Screen Time Measurement System uses Dlib and face-recognition library to locate and detect faces, it takes the webcam feed using OpenCV-python and locates the faces which are looking towards the screen, after it has located the faces, it compares them with the known faces and gets the names of people who are in front of the screen, after getting the names of faces and it starts the timer if needed, and if not needed it will do nothing after they leave the screen. The algorithm calculates their screen time.
The application needs the images of the people whom it is tracking, These images are given as input in the `data/people` folder. It outputs the time of every individual in a `log.csv` file which is updated after exiting the program, this file contains the name of the person and their time in front of the screen.
