ai shopping stylist sxsw

There’s a lot of noise at SXSW. It’s a hotbed of activity for some of the biggest names in music, film, tech and fashion, with luminaries discussing everything from their latest projects to the best ways brands can use emerging technologies. But away from the gonzo parties and lavish activations, one tech provider sits in a modest exhibit, quietly addressing a key challenge for brick-and-mortar operators: How can a physical store compete with digital counterparts that offer features like artificial intelligence-driven stylists and product recommendations?

The answer could be found in the old saying, “If you can’t beat them, join them.”

Vera is an AI personal stylist that uses hardware sensors, radio frequency identification and AI to deliver in-store recommendations. Since clothes often have RFID tags, the act of hanging it on a wall loaded with Vera’s camera and sensors allows the system to identify the item, determine the shopper’s body type and search Pinterest, Instagram and Twitter for appropriate recommendations based on how others mix-and-match the garment. The results show up on a display situated in the store, so shoppers don’t have to fire up their phones to get fashion inspiration. It’s like a real-world version of “complete the look”-style online recommendations.

ai shopping assistant stylist sxsw

With Vera, hanging a garment by the sensor triggers a fashion hunt across social media for looks based on the garment.  Courtesy photo

Vera’s creator, Moonraft Innovation Labs, aims to “make discovery easier in a more meaningful way,” said Ankit Shekhawat, head of labs.

The company uses computer vision and pattern-matching to recognize articles of clothing and match them to fashion trends online. “If you look at the garment, it has to recognize it in many ways — what kind of neck it has, for example, and what kind of pattern it has; whether it’s a top or a bottom, and what kind of colors it has. From there, it actually matches them together.”

Searches can yield a massive load of results, so Vera ranks them using factors like popularity and uniqueness, as well as the retailer’s inventory, so shoppers can check out looks featuring other items in the store.

Moonraft has signed its first partnership with a large retailer, one with upward of 1,500 locations. The launch focuses primarily on basic use cases, at least for now. But the technology is adaptable, so it can extend to the retailer’s e-commerce site or mobile app, said Sreekumar Paramu, Moonraft’s chief executive officer. The next step, said Paramu, is looking into data collection by linking Vera with loyalty cards.

“Let’s say that I’m [interested in] something,” he said. “I can actually look back at my previous purchases, for example, so I can know what I have at home actually matches with this garment.”

load comments
blog comments powered by Disqus