Inspiration
Everyone loves BeatSaber, but not everyone can afford to drop $500 on a full VR set to play the game, especially broke college students. So we decided to create a similar rhythm game called PeasantSaber - BeatSaber for the peasants.
What it does
PeasantSaber uses hand recognition technology to allow you to play a simplified version of beatsaber right from your computer's webcam. We also wrote python scripts to allow you to design your own levels given an audio file for the music.
How I built it
The game is built in python with Tensorflow, OpenCV, and some audio processing libraries.
Challenges I ran into
This project was the first time we implemented multiprocessing threads, and we ran into a ton of issues - especially with passing variables between our worker threads and main thread.
Accomplishments that I'm proud of
Figuring out how to use and structure a Queue to pass data between threads was probably our greatest accomplishment. This was a bug spent hours debugging and most online tutorials didn't help because all of our worker threads were running in infinite loops unlike most examples where the function call would end.
What I learned
We learned a lot about multiprocessing and how it is used in machine learning applications.
Log in or sign up for Devpost to join the conversation.