During the Amazon re”MARS 2019 conference, the company announced a new drone delivery program and the Pegasus robot but that was not when the surprises ended: Amazon also told the audience about an upcoming tool they dubbed StyleSnap.
The idea behind the app was most likely influenced by the low sales Amazon’s ‘private label’ has been recording : its women’s clothing line sells about 100 units per month, according to data platform JungleScout.
It’s no secret that women are more easily attracted by the visual aspect of shopping and, with the rise of social media, a lot of us get inspiration from Influencers and other people with a strong online presence, so it was only a matter of time until someone came up with an idea of how to further monetize on it.
StyleSnap doesn’t make a secret about it though: it allows high-profile bloggers of Influencers to recommend Amazon items and make money from it. The program was launched in 2017 and, according to Business Insider, some Influencers receive as much as 10% commission for the items they recommend from Amazon’s private fashion line.
If you’ve ever stressed out over replicating a look you saw in a magazine or Instagram, this tool might just make those style dreams happen.
StyleSnap is AI-powered and helps you shop around. All you have to do is open the Amazon App and click the camera icon in the upper right corner and select ‘StyleSnap’. After that, you can either take a photo of a screenshot of the outfit you like and the feature will send you recommendations for the similar items you can find on Amazon that fit what it saw in the photo.
Of course, the feature will keep in mind factors like the brand, the price range and customer reviews, to help you choose what fits your wallet and expectations easier than ever before.
Identifying what is going on in a photo is not easy for a machine, so Amazon employed the use of computer vision and deep learning to identify the items in a photo regardless if the photo showed a person just posing in front of a while background or eating a croissant in a coffee shop.
Amazon fed the neural network many images so it can learn to differentiate between your regular A-line skirt and a Scottish kilt, for example.
“To have neural networks identify a greater number of classes, we can stack a greater number of layers on top of each other.” Amazon said in a press release “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.”
Amazon didn’t say when exactly the feature will be released but it’s expected it will happen quite soon.