Styled.

Inspiration

Our inspiration comes from waking up every morning and struggling to find the perfect outfit that will make us feel our best--cause when we look our best, we do our best.

What it does

Styled. takes images and processes them with machine learning to grab just the item, whether it be clothing, accessory, or footwear. It then has these images sorted into styles to choose from. Once a style is picked, Styled. will give you your outfit, consisting of a top, a bottom, footwear, and an accessory.

How we built it

We built our project using GrabCut as the machine learning and HTML for the front end. The back end consists of primarily Python. We used a local server to launch our website.

Challenges we ran into

We first tried to launch off Github Pages, but since that is primarily for static websites, we needed to find a new server. We tried Google Cloud but it would have been very time consuming to implement it within the time frame as both of us have no experience with Google Cloud, so we decided to use a local server to demo and will work on launching it to Google Cloud at a later date.

Accomplishments that we're proud of

We are proud of the way the singular items of clothing for the outfits are produced. They have a scrapbook feel that is to fit our aesthetic and the way they are cut out also adds to the effect. Later this will allow users to add their own picture and have their items of clothing cut out.

What we learned

We learned a lot about machine learning and fitting the pieces of a project cohesively together. We found that communication is important throughout the entirety of the project instead of just at the planning and implementation stages.

What's next for Styled.

Next, Styled. will be able to take users pictures and cohesively put an outfit together for them that is trained by AI to be aesthetically pleasing, instead of just randomized. We are excited for our projects future.

Built With

• flask
• grabcut
• html
• javascript
• python