Hits To Fits

This is the landing page for our application
This allows you to upload items to the closet
This shows all the clothes in your closet
This is the final match screen

Have you ever woken up late for class, and needed to rush out the door, but did not know what to wear? Or have you ever wanted to learn how to be more fashionable? This website is for you! Just put in a song to match the day's vibes, and get an outfit suited for you!

Inspiration

We were inspired by styling apps, that used AI & machine learning to be a digital stylist. We wanted to connect music & fashion with this application! We were also inspired by the multitude of Spotify related apps, and wanted to connect this aspect of music with fashion.

What it does

This application lets users submit a Spotify song, and it will generate an outfit based on the elements of the song. The user can then give feedback on the outfit, so the AI can learn what fits the user's style.

How we built it

We built it in Python using the SpotifyAPI, Google Vision API, Google Cloud Storage, and Flask for the front end.

Challenges we ran into

We've never done AI or machine learning before, so this was a whole adventure. Additionally, we realized that machine learning takes a long time, which contributed to a wait for a match. We also ran into several challenges as we tried to get the Google Cloud APIs to work, but we figured that out eventually.

Accomplishments that we're proud of

We're proud of developing a functional web app! All of these technologies were new to us and we were all passionate about using AI and Google Cloud to build a project from start to finish.

What we learned

We learned how to develop an AI algorithm that curated to the user's color style based on previous inputs. We also learned how to integrate Python development with flask. Finally, we learned a whole bunch about the wonderful (and powerful) features that Google Cloud has to offer.

What's next for Hits To Fits

We want to be able to draw in larger datasets to match other clothing accessories as well as clothing styles. Another thing we wish to do is tie each closet to an account. This way, multiple users can use the website while having their data protected. We also want to improve our error detection.