Automation saves time and effort. We like automation. We thought: what parts of our daily lives have we not automated effectively. There are already plenty of apps for podcasts, faster text messaging, etc., but we were not able to find an app that takes our existing wardrobe and tells us what combinations go well together. We figured that an app/ website that makes these decisions for the user would be broadly applicable.
What it does
Using a database of the user’s clothing, A.I. Apparel finds the best clothes for the day through machine-learning based on factors like weather & occasion for the day (with other factors potentially being added in the future)
How we built it
We built a Node.js backend server that uses machine-learning to identify an outfit, and then presents that information on a website stylized with Google's Polymer library. The machine learning was done through training with a database of over 270
Challenges we ran into
Client-server interaction and creating a sufficient data set. And client-server interaction. And a bit more client-server interaction.
Accomplishments that we're proud of
Everything - we were focused and did a lot of work with little sleep. That's the normal life of a high-schooler, but we pushed both extremes this weekend.
What we learned
Even with programming, planning is important. And machine-learning isn't that hard…
What's next for A.I. Apparel
There are many options available: more user options, image-based machine learning, a mobile app, calendar integration. We'll probably clean up the code a bit first.