Lookify

Inspiration

We were inspired by the constant portrayals of outfits on social media such as Instagram and Tiktok. Specifically, one of our favorite NBA players, Shai Gilgeous-Alexander, would always post his outfits on Instagram and we would want to know where to buy clothes similar to his from.

What it does

Our web application allows users to input an image of a piece of clothing, and in return, it provides a comprehensive list of real clothing items that closely match the input. For each recommended item, the user can view its price, purchase URL, store information, and an image of the piece of clothing.

How we built it

We first created 2 separate algorithms on Google Cloud Vision API and LykDat API to generate a string describing the clothes that the user inputs. From there, the 2 algorithms cross-reference each other in order to create a final string that is a detailed descriptor of the item of clothing. The descriptor is then sent to a Shopping Web App API, which returns to the user a long list of clothes that are similar to the one they inputted.

Challenges we ran into

The main challenge we ran into was the fact that our API keys were public. In order to fix that, we created environment variables so that nobody could steal our API keys.

Accomplishments that we're proud of

We're proud to create a working product that searches the entire web for different articles of clothing.

What we learned

We learned many important lessons during the creation of this application such as the importance of learning on the fly with outside resources and the value of taking a break.

What's next for Lookify

We hope to continue to work on this project in order to implement new features, one of which includes creating user profiles and accounts so that users can save different outfits that they really want to purchase in the future.

Built With

• css
• flask
• googlecloudvisionapi
• html
• javascript
• python