“Eyealike VisualAd is the world’s first image and video-based contextual advertising platform for social networks and rich media websites. With a unique ability to recognize facial features, skin color, hair, gender and age as well as objects like logos or product images, Eyealike VisualAd can serve highly targeted ads based on the context of online images and videos. The proprietary approach is similar to Google’s AdWords model, but based solely on an automated method that quickly indexes, filters and classifies images and videos rather than relying on keywords to place targeted ads alongside rich media.”
I got the first paragraph straight from Eyealike’s Web site because I didn’t think I could explain VisualAd any clearer.
Eyealike’s VisualAd platform can recognize people with 90 percent accuracy – quite impressive for a computer. In a press release from October 13, the day that VisualAd was unveiled to the world, Sue Feldman, IDC’s Vice President for Search and Digital Marketplace Technologies, said, “Online ad revenue drives the digital economy, and no one has a lock on that revenue stream today. Online advertising – particularly contextual advertising – continues to soar. IDC believes that large publishers and ad networks can seize a significant share of this revenue.
Eyealike’s ability to automatically tag and classify images based on the image, not the keywords, should result in matching more ads to images, and also improve the match based on the actual image. This is an approach that could help increase the ad revenue for websites with visual assets.”
Although VisualAd is similar to Google AdWords, the amount of money that can be made is much higher. Many large social networking sites have low CPM rates, from 8 cents to 12 cents per thousand views. VisualAd’s goal is to make those same sites now worth $2 to $20 and up by increasing CPMs.
The VisualAd technology not only recognizes faces, but logos as well. If you have a pic of yourself with a Nike shirt on with the swoosh logo, VisualAd knows that you’re more likely to respond to a Nike ad displayed on your page.
When it comes to faces, the technology can spy an infant in your arms or the same infant in a multitude of pictures on your profile and assume that you need a Gerber item or a Huggies item. A financial services firm can conclude you might be interested in setting up a 529 college savings account for your little bundle of joy. Thus, seeing all of this info in a picture on a social networking site allows for more targeted ads.
This is definitely innovative – no one can argue that – but is it scary, a little too Big Brother-like?
Recent Comments