gazeMetrix Using Image Recognition Tech to Find Branded Instagram Photos

With users on popular photo sharing network Instagram going from 15 million to 80 million from January to July 2012, brands can bet that a good portion of those photos taken include their logos with people either evangelizing or criticizing products and local businesses. Which is why New Delhi, India-based gazeMetrix set out to put its image recognition technology to the test and now analyzes over 12 million public Instagram photos a day to help brands identify their logo. The team recently moved to San Francisco to participate as part of the fall class of startup accelerator 500 Startups.

The co-founders initially met at a Startup Weekend event in New Delhi, where they iterated on an original product that allowed anyone with a smartphone to take a photo of an app on a friend’s phone and download it to their own. Despite raising some initial seed funding it failed to scale, and they then turned their attention to providing video analytics, using their algorithm to analyze brand logos during sport events to measure impressions. Finally, the company turned its attention to social media, and with the proliferation of photos and an image recognition technology that determines unique brand signatures in photos, it finally found its niche.

“With social media, it’s really exploding, we get 12 million pubic photos every day, and it’s growing exponentially. We’re looking at all these pictures, identifying brands in them, and letting brands use these pictures as an engagement tool back to the users and with their friends. Brands seem pretty interested in getting to know these people who are sharing their pictures,” co-founder and CEO Deobrat Singh said in an interview.

The company’s technology uses an algorithm that breaks down the unique characteristics of a brand’s logo, everything from the corners, shapes, lines, shadows, and colors, to create a brand signature. From there each photo is processed using what Singh termed ‘fuzzy matching,’ which means that even if the logo in the image is only partially showing, is on a wrinkled t-shirt or any piece of clothing or anywhere else, it will still pick it up and match it to the brand. What brands can then do is aggregate all the images containing their logo and eventually will be able to interact with those who uploaded the photos to boost brand engagement.

“We’re going to actually launch in the next couple of weeks a private beta, invite only, to allow brands to actually start using the engagement tool wherein they can actually see these pictures and share on their Facebook fan pages with the users tagged, the user who actually took the photo, so that they’ll get featured  and these users are very likely to share with their friends and generate a viral loop,” Singh added.

GazeMetrix first plans to open up the engagement dashboard in private beta on an invite-only basis to a handful of brands but envisions charging a monthly subscription fee when they’re ready and done with testing. The company is also debating whether it will charge brands a separate cost per impression for letting them share the photos or build that into the subscription fee.

Although other companies like VenueSeen launched a tool a few months ago to help brands run DIY Instagram campaigns by aggregating images based on hashtags, gazeMetrix is looking to match brands with their fans based on the images themselves. It capitalizes on the fact that Instagram users may not add any comments or hashtags but still include a brand in a photo which would go unnoticed if they only searched via text and hashtags, a phenomena brands are also encountering with other visual and image-based networks like Pinterest.

The company plans to eventually roll out its technology to other social media platforms and plans to approach Pinterest, Facebook and Twitter. With 500 Startups demo day in a few months, it remains to be seen if the company can build out its engagement platform as it starts bringing on brands, and what value add it will be able to provide them. With other startups like Stipple and Crowdsend trying to create a business around annotating images and providing information about their contents, gazeMetrix is playing in a space that’s likely to only get busier as image-based social networks proliferate.



Humayun Khan

Humayun Khan is a Senior Writer and Analyst at BetaKit. A marketing graduate with honors, Humayun's work experience spans the fields of consumer behaviour with noted contributions in an academic paper published in the Journal of Consumer Psychology and market research consulting having coordinated projects for a major financial services client at Decode Inc. More recently he was involved in business strategy as a Business Analyst for an equipment rental outlet and prior in the National Marketing Department at Ernst & Young LLP. He is passionate about emerging and disrupting technology and its ability to transform and create entirely new industries.

0 replies on “gazeMetrix Using Image Recognition Tech to Find Branded Instagram Photos”