A face-recognition activated alarm clock based on Kivy, PyTorch and OpenCV.
This is just a pet project.
This project was created for participation in SBUHacks '20 because the author thought it would be fun to record your morning faces.
If your daily morning routine involves looking at yourself in front of a mirror within 5 minutes of waking up, like the rest of us, chances are that you have a pretty solid idea of your own bottom line.
A research conducted in 2018 between 400 male/female participants generally aged around 18-40 investigated the relationship between the subjects' attitude towards life and the first image they see of themselves upon waking up every morning. Strong causality was found in the results, in that subjects who woke up to the best portraits of themselves were generally happier, compared to those who woke up to the worst portraits of themselves. General confidence, mood levels and productivity were part of the metrics that defined 'happiness'.
Disclaimer: The above is completely fictional, albeit makes much sense, am-i-rite? 🤷🏻
This awkwardly named FaceRecoAlarmClock that we eventually renamed to MorningFace helps people improve their "wake-up faces" by embarking on a gradual process of consistent self-portraiting upon waking up. And what better way to do that than to integrate it into an alarm clock?
The alarm clock in MorningFace works just like any other, with a catch: no turning off until you stand in front of it and pose for a hearty bedface portrait!
Just clone the repository, and run the app from terminal/cmd by doing python AlarmClockApp.py. The ML-integrated smart alarm clock will take care of the rest. Pretty self-explanatory right? Now go see your Morning Face get better and better!