New Balance Uses Machine Learning With Digital Poster To ID And Reward NYC Fashionistas

The footwear brand New Balance did an interesting machine learning-driven digital signage piece on the streets of New York recently, putting up a screen and sensors that scanned approaching pedestrians and flagging those people who weren’t dressing like everyone else.

The “activation” – run for a day during New York Fashion Week – used camera-based pattern-detection to look in real-time at the colors, patterns, shapes, styles and other features of people’s outfits, and then pointing out “anomalies and exceptions to the norm.”

New Balance had street marketing crews stopping people if they triggered an “Exception Spotted” notice on the screen at a Soho store. Those people were rewarded with a pair of the brand’s sneakers.

“The idea is to celebrate people who go left when everybody else is going right,” said Allie Tsavdarides, director of global marketing at New Balance, of the “Be the Exception” campaign. “During a week where there is incredible emphasis and excitement around new trends and fashion, New Balance wants to celebrate individuals who are expressing themselves in independent and distinctive ways.”

The exception spotting came in the week of a team from the marketing agency VML, in the run-up to fashion week, running around the city collecting baseline data about fashion trends in order to feed the system.

Here’s a video from Mashable about the program. Not sure about the part at 1:30 where the agency guy casually says “we have some neural networks running in the back here…” Don’t think it is quite that simple, but then, I’m no AI expert: