Snapchat Patent: Serving Ads Based on Object Recognition
Details have emerged about a Snapchat patent that would serve users sponsored filters, advertisements, and coupons based on the photos and videos they take.
While the patent was first filed in January of 2015, the United States Patent and Trademark office released the documentation last week.
The Snapchat patent application offers the example of a user taking a picture of the Empire State Building:
“For example, if the user takes a photograph and an object in the photograph is recognized as the Empire State Building, photo filters associated with the Empire State Building may be provided to the user for use with the photograph. In this example, a picture of the Empire State Building may use a King Kong filter that would place the giant ape on the Empire State Building in the photograph at different perspectives. Therefore, a picture of the south face of the Empire State Building might see King Kong’s back, while a picture from the north face might see King Kong’s face looking at you.
In this example, provision of the King Kong filter may also be constrained by a geo-fence (e.g., geographic boundary) around the area in New York including the Empire State Building. Of course, further Empire State Building-themed photo filters may also be presented to the user. The presentation of the photo filters to the user may be in response to the user performing a gesture (e.g. a swipe operation) on a screen of the mobile device.”
The filing goes on to note that the same technology could be applied to videos. It envisions a bidding system in which different advertisers would pay to have certain objects associated with their content.
Aside from the sponsored filter angle, the Snapchat patent also notes that the same technology could be used to offer coupons, restaurant menus, and more.
Business Insider has an easy-to-read version of the Snapchat patent application in full. The Verge reminds us that this is, as far as we know, only a patent application, but it’s still amazing to consider all the possibilities that image recognition technology offers.