Inspiration
We wanted to explore something no one had a lot of experience in. Computer vision seemed like a domain that would be considerably useful based on its various applications. We specifically chose sleep detection as a topic as sleep deprivation is a very big problem.
What it does
Finds out the number of times a person blinked and the total number of frames when their eyes are closed. This in itself might not be too useful, but it can be used to build a base for extensions such as online meeting platforms, or hardware webcams
How we built it
Used OpenCV and Dlib with python programming.
Challenges we ran into
Differentiating when to increase the number of frames and the number of times a person blinked was a major challenge. Along with python indentation ;)
Accomplishments that we're proud of
Building a version that does not have a lot of bugs
What we learned
Definitely learned a lot about artificial intelligence and computer vision.
What's next for Drowsiness Detection
We are planning to upgrade it to include the time for which a person blinked in seconds rather than the number of frames. We would also like to do some more work to make it more accurate.
Log in or sign up for Devpost to join the conversation.