Inspiration
Our team came into SwampHacks with a vague desire to use AI elements in a game. During the sponsor presentations, we saw all of the Google Cloud APIs and were inspired to use the Vision API to control the game. We brainstormed ideas of old games we were all familiar with and eventually settled on Doodle Jump. After that, we realized that the Vision API is more for facial detection within images instead of live-video feed, but we decided we all liked our initial idea and stuck with it. After a bit of searching, we found OpenCV to be compatible with our goals and settled on doing the project with PyGame.
What it does
Gator Jump! is similar to the classic mobile game "Doodle Jump", but we have a more advanced game control that brings more fun and excitement than any other ordinary game. You will need a WebCam to play in our Facial Control Mode because we are using Facial Detection! You will be able to control your character's fate by moving your face alone with him! Come check it out!
How we built it
We built it with Python using Pygame and OpenCV.
Challenges we ran into
The most difficult challenge we ran into is to constantly keep track of the user's face in games. With our original design, it was easy to lose track of face when the user tilts their head. After Implemented some math functions, we are able to keep track of the user's face with angles tilting from -25° to 25°. It was difficult to maintain a consistent framerate in our game with facial detection on. But we are glad we made it work!
Accomplishments that we're proud of
It's was our first hackathon. We are proud that we accomplished all of our goals. We feel great about being able to deliver an exciting game with many elements that we never knew of before.
What we learned
Two of our members are first-time Python users and it feels good to learn a powerful language. We also learned how to use external libraries, how to use and implement facial detection, and most important of all, we gain friendships.
What's next for GatorJump
We will keep working to improve the constancy and accuracy of the facial detection. We will also work on the game to bring more exciting elements to increase in-game experiences.

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