Inspiration
A lot of us suffer from not being able to focus properly on homework and getting distracted, so we figured it would be a good idea to help people who also have this same issue!
What It Does
Our carefully hand-crafted machine helps "encourage" the user to spend more time looking at their computer than down at their phone!
How We Built It
We used two Machine Learning Models, one taken from OpenCV's open source public library to track the face and the second one created and trained in-house to check the user's eyes. If it seem's like the user's eye's are closed for too long or they seem to be unfocused, we help "encourage" them back on their work.
Challenges We Ran Into
The model to detect eyes was very difficult to create. We went through several iterations, changing the output from a binary sigmoid activation output to finalizing it with just a softmax output with 2 options. We also tweaked the architecture before finally settling on the CNN with 16 -> 8 neurons.
Accomplishments That We Are Proud Of
We were able to fully integrate hardware and software with this project after overcoming many struggles. As stated, training the CNN from scratch was very difficult, but also fully creating an automated water-gun shooter, which uses motors and arduinos to encourage the user to keep focusing, made this task a simultaneously challenging and fun one.
What We Learned
We learned how to work together and use Source Control Methods effectively. This was all of our first times hacking in a hackathon and coding in a team environment, thus learning to use SCM's was a huge learning curve.
What's Next for LockIn
We're looking to really create a sense of encouragement in the user. Obviously a squirt of water won't fully encourage them to continue working, so we've been looking to expand into the taser industry to be able to fully encourage our users.
Log in or sign up for Devpost to join the conversation.