Inspiration
The inspiration for building this model came from the Iron man movies. I have seen Iron man movies a lot and got inspired by the AI used there in the Iron man suit. Once I saw a scene where Tony stark gives instruction to his AI model (aka JARVIS) to scan the whole room and find the thing necessary for future use. In that scene, the AI scans the entire room but in my small model, it will browse and detect the movements of your hands and join it via lines.
What it does
After running the model it will pop up a window in your PC basically a camera window in front of you and you have to place your hands in front of the camera and the model will detect and create a pattern of the movements. It will create 21 points on your hand and create a pattern by joining them.
How we built it
I have built it using Python and some libraries like Mediapipe and Opencv.
Challenges we ran into
The main challenge for me was to make sure that I had a clear understanding of the library's special functions used in order to run the model successfully. I was familiar with the libraries one year ago when I was in my freshmen year but when I try to implement them now I forgot how to implement so this was my biggest challenge I revised and practice the syntaxes and functions got some errors and stuck a lot throughout the process but ended up by figuring out those problems.
Accomplishments that we're proud of
To be quite precise I am not proud of what I have built because I have to deploy the program on a website using Flask and want to run the program live with one click and in between the process ended up with a lot of errors and incompatibility issues which leads to ended up my time in solving those errors and could not accomplish the whole project that I wanted to. But still, whatever I have built at least I have tried and I am proud of it.
What we learned
Things I have learned
- Time management
- Troubleshooting
- Package management and dependency issue resolvement
What's next for Detectovert
Some exciting surprises are in line
- Deployment on a live server
- More features
- Some integrations that I can't do due to limited time and some more surprises.
Log in or sign up for Devpost to join the conversation.