Inspiration
As a team of artists and writers, we were interested in how we could process visual arts or creative writing through machine learning, or specifically, the fields of image and natural language processing. We originally considered training a network to be able to detect art forgery, but due to not finding relevant datasets, we decided on converting images to a Monet-style painting.
What it does
Upload an image of any size and of .png, .jpg, or .jpeg file type, and click the 'Monetify!' button to convert it into an image that resembles Monet's painting style.
How we built it
The frontend was built with HTML/CSS, and the backend and ML portions were done in Python.
Challenges we ran into
In the frontend, we had a lot of trouble getting the visuals to display nicely on different screen resolutions, especially since it didn't seem to scale properly with some resolutions. In terms of running the application, we also ran into issues with using Flask in Ubuntu because it kept giving an error regarding a specific module, but we fixed this by switching over to a different tool to run Flask.
Accomplishments that we're proud of
We're proud of the simple, yet aesthetic UI of the front page as well as getting the model to work separately despite not being able to hook it up with the main application.
What we learned
ML is an extremely complex yet fascinating topic, capable of all sorts of feats. In addition, most of us were unfamiliar with Python, so we got to gain some more experience in that.
What's next for Monetify
Firstly, getting it to work with uploaded images. We also thought allowing the user to view past submissions they made within their ongoing session would be a neat feature!
Log in or sign up for Devpost to join the conversation.