Inspiration
Some inspiration is living in Vegas and seeing all the wonderful mathematical applications we might not expect, even if it is right in front of us.
What it does
This project implements a real-time gesture-controlled slot machine using classical computer vision techniques.
How we built it
This program is built in python using NumPy and OpenCV
Challenges we ran into
There are tons of challenges I ran into. First the UI was incredibly hard, the centroid math stumped me, initially even getting setup with OpenCV and using a camera in python was a hurdle and many more. The math for tracking and thresholding was harder than I anticipated.
Accomplishments that we're proud of
I'm extremely proud to output a function computer vision program using real time calculations and display, and combing my interest in computer science, computer engineering and applied mathematics.
What we learned
I learned a ton of things ranging from classical computer vision, Region-Of-Interest filtering, centroid mathematics and tracking, frame differencing, NumPy and creating UI's in python, event retriggering bugs, git version control/merge conflicts, transition-based scoring, thresholding, indentation bugs, but ultimately shipped something stable.
What's next for Vision-based gesture-controlled slot machine
There are many steps to take forward. The primary focus is fine-tuning the functionality and eventually changing to a neural network (R-CNN most likely) that can better detect the object (hand) and warrant a much more effective output via better calculations. Beyond that, making the UI/UX much prettier and efficient is a goal certainly.
Log in or sign up for Devpost to join the conversation.