Inspiration
We saw some people talking about a very similar idea on Reddit, but they didn't know how to implement it. Our team thought we could implement it, so we gave it a shot! We also think it's a really cool idea, for example, it could be implemented in an online game for randomly re-rolling your flag.
What it does
Our project auto-generates new flags based on the flags we used as our training for the machine learning, and responds to user inputs in generation
How we built it
We used TensorFlow and an autoencoder to build the model.
Challenges we ran into
We ran into a few challenges. The largest would be a challenge we still have not addressed, that of the model outputting a flag that is more blank than we would like. We also faced the challenge of having less time to build our project. We only began on Saturday evening.
What's next for Autodecoder Flag Generator
Ideally, we would like to use a GAN network for in outputting different images. We also need to fix the quality of the flags being output.
Built With
- python
- tensorflow
Log in or sign up for Devpost to join the conversation.