myCloset.io

Inspiration

Morning routines are the worst, and picking what to wear in the morning makes it that much harder. We decided to build a solution to help us make the best outfits, making our mornings a little easier.

What it does

myCloset takes a piece of clothing as an input. We detect the main colours and patterns of that article of clothing and start suggesting what other pieces of clothing might work with the inputted article after going through our colour matching algorithm.

How we built it

The core of myCloset is a custom made colour matching algorithm which we developed with different fashion theories regarding colour.

Challenges we ran into

We tried implementing the Google Vision API to automatically detect which article of clothing and colour the user was inputting, but we ran into problems with Android Studio as we weren't able to save the pictures taken of each article of clothing.

Accomplishments that we're proud of

We're proud of the suggestions our algorithm is suggesting. Our code wasn't working until 4am last night, and we pushed through the rest of the night to perfect it.

What we learned

Devansh - This is my first hackathon so I learned a lot about OOP, Javascript, how to use APIs, and the importance of sleep Prince - The science behind colours and why some colours work well with others, and not so well with others. Paul - I learned about how to parse the JSON responses from APIs. Tolu - As a web dev, I learned a lot about mobile dev and how the Google Vision API works.

What's next for myCloset.io

Next up, we plan on successfully implementing the Google Vision API to improve UX. After that, we plan on implementing a machine learning algorithm so myCloset can learn what the user's sense of style is, and better tailor suggestions.