I’ve been devoting a good amount of time over the last two weeks to my new girlfriend, a real cutie who is a personal stylist of sorts at Urban Outfitters.
She’s pretty, knowledgeable, and can make anyone look fantastic. Her name is Snap & Match … and she’s an iOS application that I’ve developed for my summer intern project with Billy as the backend developer and Natalie as the UI designer. Her codebase is all Swift/Objective-C code (no storyboards for me!). It was truly a pleasure getting to know iOS development through her.
Sad analogy aside, I’m actually really excited to show you this app. It’s something that fifteen-year-old me, who only wore the colors navy and white, would have appreciated. They do say to be the person you needed when you were younger, so…
Color-wise, people tend to stick to their comfort zones when shopping for new clothes. They want to experiment with different colors, but aren’t sure where to begin. Snap & Match makes this process easy by taking into account the colors you already own, and suggesting clothes that could compliment items in your existing wardrobe. The app uses color theory to make its matches.
Currently, Snap & Match will give you Urban Outfitters clothing suggestions, and matches complementary colors. For example, if you’re looking for new shoes and snap a picture of your purple dress, the app will give you suggestions for yellow shoes from UO.
First, pick the type of clothing you’d like to buy.
Take a photo of the clothing you’d like to match to …
… and get back your results! It’s that simple. Each item on the list links to its product detail page on the Urban Outfitters site, and you can shop directly from there.
How It Works
We utilized the Google Cloud Vision and Urban Outfitters Catalog APIs in Snap & Match’s backend. When users click “Use Photo”, the Google Vision API detects the dominant color of that image. We determine that color’s complementary color and pass it, along with the user’s selected clothing category, to the Urban Outfitters Catalog API, which gives back clothing suggestions.
Will Snap & Match be available in the App Store?
The answer is maybe! As of right now, Snap & Match uses UO’s proprietary libraries and APIs, and would need special permission for distribution. I’ll keep you updated.
The Experience // Future Improvements
In the future, Snap & Match will have more match options based on other color theories, such as analogous, monochrome, triad, or split-complementary. It will also suggest clothing items from other brands.
I’m immensely proud of the work my team has done and grateful to all of the advice, help, and encouragement we got along the way. I know it’s only been two weeks, but I’ve learned a lot from this relationship 😉 I intend to make search functionality smoother and keep adding features to the app.
If you’re interested in the personal insights I had during development, I’ll be doing an in-depth post about my experience interning at URBN, including what it was like to build Snap & Match, later this week.