Inspiration
More and more people get diagnosed with ADHD every year which puts them on a disadvantage on the educational field, and with many students cramming for the upcoming finals, we decided to create a web app which helps with studying.
What it does
It detects when the user starts to fall asleep/doze off and outputs audio and light waves and to wake him up. There is also a stopwatch which records how long the user has been studying for.
We designed our website using a minimalist design to reduce distractions with shades of blue which further promotes concentration.
How we built it
We used HTML5 and JavaScript to create the web app and the backend was done using Tensorflow and Opencv. We found a database of pictures with people blinking and used it to train our Multi Layered(3) Convolutional Neural Network for 3 epochs which gets us a model with and accuracy of 93.69%.
Challenges we ran into
The timeframe of the Hackathon. We didn't have enough time to use a cloud computing service to fully train our model and link it to our website.
What's next for ConcentrateAI
Built With
- convolution-neural-network
- html5
- javascript
- keras
- opencv
- python
- tensoflow

Log in or sign up for Devpost to join the conversation.