Inspiration
We were inspired by the website https://thispersondoesnotexist.com, deepfakes, and how realistic images produced by GANs (Generative adversarial networks) can be.
What it does
We created a website where people play a game where they are given one image that is real and one image that is produced from a GAN. Can the user detect what's the GAN or not?
How I built it
We built it using React front-end that pulls images from a real image dataset and a GAN-generated dataset.
Challenges I ran into
We were originally going to customize our own GAN, but complications arose so we decided to have a fun and easy application using pre-existing GAN architecture.
Accomplishments that I'm proud of
I am proud of my team members of understanding the concepts of a GAN and putting it into use.
What I learned
An application of GANs! Also, some of us had never used React or done front-end before.
What's next for Gan Game
Who knows where GANs can take us next?
Log in or sign up for Devpost to join the conversation.