Inspiration
It's easy to feel like an alien when you don't understand what face to make and when. Alien Interrogation took inspiration from the nuances of social interaction, and how some people, especially those with neurodivergence, struggle to understand them.
What it does
Alien Interrogation involves a simple gameplay loop, using facial recognition as an input for gameplay. You are being interrogated and shown photo stimuli. Your job is to make the right face and blend in as a human.
How we built it
We utilized OpenCV and GameMaker to detect live facial recognition from the user. First, we utilized Copilot within Visual Studio Code to create a python program and set up a webserver. Using the OpenCV webserver and MediaPipe, we were able to return the shape of the user's face as a JSON. In GameMaker, we requested the webserver for the result, and this allowed us to create an interactive facial recognition experience in Alien Interrogation.
Challenges we ran into
We persevered through conflicting GitHub commits, being abandoned by a team member mid-hackathon, and installing dependency after decency for OpenCV. The process to create Alien Interrogation was difficult yet rewarding. A major challenge we overcame during development was handling facial recognition and understanding different emotions. We switched detection models after prioritizing facial geometry over machine learning.
Accomplishments that we're proud of
Alien Interrogation boasts quick, consistent facial recognition, a vibrant atmosphere, and lots of love put into its development. We are very proud of the project we have created.
What we learned
We became comfortable using GameMaker and requesting live responses from a webserver.
What's next for Alien Interrogation
We hope to detect more emotions, specifically anger and disgust. We also hope to flesh out the story more and explore the vibrant atmosphere.

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