Inspiration
I had been personally affected by the incorrect marking of attendance and had faced the long and tedious process of fixing it with the school. I took this problem that I encountered and decided that I wanted to create a possible solution to address it.
What it does
MarkMyPresence is an automated attendance marking system that is powered by Machine Learning and AI techniques. It makes attendance marking in schools automated through facial recognition and ML, matching students that are present and marking them present. Common errors with attendance management are avoided and saves time & hassle for students and teachers
How we built it
We used PictoBlox, a ML based environment to code and run our project.
Challenges we ran into
We had some difficulties with defining and training the ML model. We also had some trouble with the accuracy of recognition since we didn't train the model for a very long time.
Accomplishments that we're proud of
We are proud that we were able to step out of our comfort zones and pick up to code in ML which neither of us had experience in. Even though the project is still a work in progress, we are proud to have made a working simulation with the time we had.
What we learned
AI and ML can be used to solve complex pattern recognition problems and define solutions. We also learned how to code with ML for the first time.
What's next for MarkMyPresence
*Partner with commercial ML solution providers like AWS *Model accuracy *Use more training steps (up to 50,000) to achieve up to 100% accurate results *More student samples *Fully working demo with live and static images *Expand to school/multiple schools/within school district *Reports and alerts for teachers and administrators *Can be expanded to scan for facial emotions and notify support counsellors to take further action
Built With
- ai
- ml
- pictoblox
- python
Log in or sign up for Devpost to join the conversation.