At the MARS 2019 conference in Las Vegas, attendees were introduced to Amazon’s StyleSnap, a visual search AI app that can identify the clothing featured in digital images and find them (or similar items) in Amazon’s online stores. Built using AI and deep learning technology, the app can classify articles of clothing in seconds and provide a way for its users to purchase their favorite looks instantly.
This powerful new capability, known as visual search, is growing increasingly popular among Millennial eCommerce app users, with 62% stating that they look for visual search capability in the mobile shopping apps they use.
In introducing StyleSnap, Amazon is hoping to build on the popularity it enjoys with Millennial, incorporating the app into a broader campaign featuring Instagram fashion influencers and exclusive lines of designer clothing that it intends to release through a shopping experience called The Drop.
While StyleSnap may represent only one part of the overall campaign, it represents a large technical effort on behalf of the company. CEO Jeff Wilke highlighted this at the MARS conference by remarking, “the simplicity of the customer experience belies the complexity of the technology behind it.”
The effort was further described by Amazon in its own blog:
To have neural networks identify a greater number of classes, we can stack a greater number of layers on top of each other. The first few layers typically learn concepts such as edges and colors, while the middle layers identify patterns such as “floral” or “denim”. After having passed through all of the layers, the algorithm can accurately identify concepts like fit and outfit style in an image.
We must go one step further, however – feed-forward neural networks will stall and eventually degrade after a certain number of layers have been added. This is known as the vanishing gradient problem, where the signal from the training data is so spread out between layers that it is lost entirely.
Amazon uses residual networks to overcome this problem, as they use shortcuts to allow the training signal to skip over some of the layers in the network. This helps the network learn basic features like “edges” and “patterns” first, and then focus on complex concepts. A unique method developed by Amazon researchers allows the network to learn new concepts while also remembering things it has learned in the past – this is critical for enabling StyleSnap to work through large volumes of data effectively.
While StyleSnap and its underlying technology was developed by Amazon, visual search has been incorporated onto the strategies of other large online retailers such as Ikea, ASOS, Wayfair, Argos, and Neiman Marcus, all of which have now deployed proprietary products of their own, and this trend is expected to continue.
IndustryArc recently reported that the global visual search market will increase to over $865 billion by 2022, and Gartner estimates that by 2021, early adopter brands that redesign their websites to support visual and voice search will increase digital commerce revenue by 30%.
As of 2020, new third-party tools have been introduced that are bringing the time, effort and cost of including visual search technology down to an affordable level for all business sizes. And yet, the number of small to medium-sized retailers that plan to invest visual search technology for eCommerce remains at a mere 8%.
To stay competitive and meet the expectations of their customers, small and medium-sized companies should be preparing for the coming visual search wave. Visual search is becoming more and more sophisticated and user-friendly, making it an increasingly powerful tool for brands and e-commerce platforms.
Would you like to know more about current eCommerce that might affect your business in the next few years? Do you have experience with visual search technology, or have you begun the process of optimizing your eCommerce site for visual search?
Using EndPoint Modeling, we have helped hundreds of businesses to achieve their custom software engineering goals while maintaining a 100% satisfaction rating on project delivery criteria. Our eCommerce experts can help you to incorporate the latest technologies and achieve your goals on a guaranteed budget.