We wanted to create a webapp that could help people around the world to find similar artworks to those that they liked.
What it does
smART is a webapp that uses a machine learning algorithm to recommend new art! You can import your own art or click any of the suggestions to have similar art shown. Our model can recognize all types of art - from sculptures to paintings to sketches; then recommend similar ones! Whether you're looking for inspiration or simply want to browse, this is the place for you.
How we built it
The frontend of the webapp runs on
bootstrap. The user can upload an image or open a suggested image, which calls a
tensorflow model on the backend that generates a list of suggestions. All of this is hosted on
firebase so that it's accessible anywhere!
Challenges we ran into
- We had some trouble figuring out how to use firebase.
- It took us a long time to figure out how to upload user images and display the image that the user uploaded.
- Flask was overriding the css file we coded, so to took some time to fix that.
Accomplishments that we're proud of
We have a machine leaning algorithm that can output a similar looking image according to an image the user uploads. Our website is simple, with only one button to upload an image of an artwork and a suggested artworks section.
What we learned
Splitting up tasks well saved our bacon as there were timezone differences and some other problems. We learned how to allow user to upload their own images and display the images they uploaded.
What's next for smART
A refresh for the frontend and a code refactor would both be great to do.
Log in or sign up for Devpost to join the conversation.