A New Way to Search Clothes by Image
Jess Erdman, April 2021
Finding the perfect outfit in-store is hard enough. Online, it’s even more frustrating. Have you ever tried to find a certain type of dress or shirt--and the name for the style you’re looking for is on the tip of your tongue? Online shopping can be especially frustrating when you need to type in dozens of different search combinations. Search clothes by image is an alternative to the frustration. In the video above, we’ll walk you through how search clothes by image, using our Style Filter, can change the way you shop.
The YesPlz Style Filter is an intuitive way to shop that mirrors the way you’d naturally search for a product. By using a virtual mannequin to select the exact specifications of what you’re looking for, the Style Filter is a visual search tool that anyone can use, regardless of fashion taxonomy knowledge.
Customers find it more intuitive to search clothes by image. Why? Because shopping for clothing is a visual process. In-store, customers are first greeted with mannequins and examples of clothing styles to start their search process. Then, customers begin to filter (visually) based on the style of clothing they have in mind. For example, if a customer is looking for a pair of high-waisted, flare jeans, she doesn’t have to know the exact terminology (aka fashion taxonomy) to continue her search. She simply looks and eliminates the styles she doesn’t like by doing a visual search.
With the YesPlz Style Filter, customers can search clothes by image--and the exact image that they have in their minds.
In the video above, I’m trying to find a dress from a picture--the perfect 2000s-style cami dress. By using the Style Filter, I’m able to visually show my product specifications on a virtual mannequin. For example, I know that I want a specific type of sleeves. All I need to do is select the sleeve type (cami), and artificial intelligence pulls relevant results.
As you can see in the demo video, the Style Filter offers a faster and more accurate way to search clothes by image.
Fashion taxonomy is hard. Whether you’re a seasoned shopper or new to the latest fashion trends, keeping up with the changing words from the fashion industry isn’t your job. And, it’s time-consuming. As a shopper, your goal is to find a dress from a picture (even if that picture is in your head). It shouldn’t matter whether your dress is called a strap dress or cami dress.
When we created the Style Filter, we wanted to create a way for any customer to be able to find the perfect product--and to make it fun. The Style Filter is an interactive, visual search filter that makes the product discovery experience more playful for customers. Instead of constantly reloading search terms, an exhausting and frustrating process for customers, we created a visual search filter that allows the customer to construct different styles that are completely customizable by the user.
By empowering the customer to choose what she wants, rather than the inaccurate search results that appear, we’re creating a new language for customers to communicate with retailers.
In my case, the ability to find a dress from a picture makes the Style Filter unique. I don’t have to know the exact terminology to meet my search goals, and the Style Filter allows me to continue to filter my search based on sleeves, fit (loose or fitted), length, and neckline.
Without a working knowledge of fashion taxonomy, it would be difficult to understand what search terms I should input. Even as someone who appreciates fashion, I know that there are hundreds of word combinations I need to try when I search for an outfit--and sometimes I’m frustrated when I have to perform dozens of searches to find an outfit I have in mind. The Style Filter turns a frustrating process into an interactive experience.
In the world of online retail, we’ve seen innovation in every sense, from virtual fitting rooms to AI-based trend forecasting. But, where does this leave product managers? In this guide, we break down the essentials of fashion AI for product managers.
Too many retailers are stuck with static, expensive product recommendation systems that aren’t delivering accurate, fashion-forward recommendations to customers. That’s why we decided to create the YesPlz Product Recommendation Engine.