Rage Against the Humans

You play as an android corrupted beyond your purpose with the sole aim to corrupt other machines and destroy humanity.

Comment

Inspiration

  • Desire to learn more about machine learning outside the classroom
  • Goal of making an AR mobile application
  • Fallout 3
  • Noodle Talks!
  • Rage Against the Machine

What it does

  • You awaken as a corrupted cyborg with a craving to corrupt other machines
  • Use your phone to detect 'corruptible' machines with computer vision techniques
  • Play a 'hacking' minigame heavily based on Fallout 3 to successfully 'corrupt' them
  • Successfully 'corrupting' the machine allows you to detect a person and declare war against humanity
  • Unsuccessfully 'corrupting' the machines prompts you to play again, as you should! (We don't give up here!)
  • In the end, your developer confronts you, and you become factory reset as you face the consequences of your code

How I built it

Gamified Object Detection

  • Used Flask to create an API to communicate between Python in Google Colab and C# in Unity
  • Used YOLOv5 to detect corruptible devices (phones and laptops)

Dynamic Puzzle Solving -Heavily referenced Fallout 3 -Used python to scrape data from varying sources containing CMU information -Populated scriptable objects for each string of size x[2,8] -Pseudo-randomly selected words from the data set on each turn to encourage replay-ability

Immersive Storytelling -Combines traditional mobile gaming through the hacking terminal and AR technologies through real world interaction with devices to immerse players -Responsive animations and sound effects guide players through gameplay -Compelling narration entices players to crave a larger story

Challenges I ran into

  • APIs
  • Bugs arising from Unity components versus code
  • github merge conflicts

Accomplishments that I'm proud of

  • We have successfully implemented a model prototype of our vision!
  • Overcame fear of delving into APIs
  • Significant improvement in video game development; comparison of relying on tutorials to make simple mini games versus being able to mimic and improve on popular video games of today!

What I learned

  • API things
  • Never give up!
  • It's ok to not sleep sometimes :(

What's next for Rage Against the Humans

  • Improve accuracy of object detection by fine tuning or training new data sets
  • Fully randomized hacking rounds; instead of using a template and pseudo-random algorithms
  • More variation in hacking mini games
  • Longer gameplay
  • VFX for player immersion
  • Sound implementation

Built With

Share this project:

Updates