Google Glass Taken Another Leap Into Application Development With Clothes Recognition.

, March 9, 2013

 

Google-Glass-photo_610x306

Will advertising take a Google’s Glass technology to the next lever?  A new technology in Google Glass will be able to detect people by their clothes.

A company called InSight systems, who also funded by Google, will be able to use their new technology to be able to identify people uniquely via their smartphone or using Google’s Glass. As you interact with people up close, Google Glass can inform you of personal contact information of the person you are in front of.

Google has a goal in mine and that is for people using Glass to be able to find friends and family in a variety of different places like Stadiums, Airports, and Concerts. Of course, there is benefits or drawbacks, depending on how you look at it to using this technology.

Google will create a “visual fingerprint” depending on the person’s clothes or what they might be wearing at the time. If the person changes clothes or what they are wearing the InSight technology will count that person as a new person thus changing the “visual fingerprint” or identifier of the individual. So this could mean that after one day you might have to reset, or at least know that the long-term affect is secured and more private for the individual as the system might not recognize him/her at a later time … For now.

How will this work?

The application will take a series of of pictures to identify different positions and areas of the individual. The application will the create something called a “Spatiogram” to identify shapes, colors, and so forth that is unique to the individual. Google’s Glass will then do the rest and put a “visual identifier” on that person which can then be viewed as a contact.

Facial recognition systems of course are not a new thing. But the InSight system is designed for those certain moments where you can’t see  a face, or the individual is far away, or has his/her back turned.

The system is still in the early stages, but using 15 volunteers, InSight was able to identify people 93 percent of the time.

Associate professor, Srihari Nelakudit, of computer science and engineering at the University of South Carolina,  developed the InSight software along with three colleagues at Duke University.

About the Author:

Limited Time Offer: Get up to $300 in Facebook Ad Spend on Us!LEARN MORE
+