Online Image Recognition Technology Can Target Ads

A Seattle-based Web marketing services firm Eyealike has introduced VisualAd, an automated system for capturing online visual data and targeting ads on social networking and other Web sites.

This technology enables marketers to segment advertising messages to match specific kinds of online images and videos, which consumers have uploaded to Web pages.

VisualAd technology uses artificial intelligence to recognize facial features, skin color, hair, gender and age, as well as objects such as logos and brand name products. Such information is filtered and classified for targeting ads.

Advertising is targeted on the basis of inferences made from analyzing online images. For example, an inference can be made that a person holding an infant is likely a parent, or depending on their age, a grandparent. In this scenario an ad for a baby product can be targeted online.

“Keywords and profile data tell us a lot, but imagine being able to serve up relevant ads to consumers based on facial or physical attributes, or specific objects in images such as Gucci sunglasses or logos like Nike,” said in a statement Greg Heuss, president of Eyealike.

This technology can be used with any Windows or Linux capable server for targeting advertising to computers or cell phones, according to the company.