New Google Glass app InSight takes the experience of spotting your friends to the next level, as it recognizes your peers by their clothes and accessories – matching specific color patterns and textures to the user.
This new human recognition software makes it impossible to lose your friends in a packed place, such as an airport, concert hall or shopping center.
The app creates a ‘fashion fingerprint’ for every friend based on the outfit they are wearing including clothes, jewelry, glasses, etc.
‘Fashion fingerprint’ is generated by snapping pictures of the user and creating a file of that image called a ‘spatiogram’, which calculates patterns and spatial distribution of colors on the user and analyzes the information to identify that particular person later on in the distance or from a different angle.
The new system is being partly funded by Google and was presented at the HotMobile technology conference last week, while currently still being developed by the Duke University in North Carolina.
So far the app has been tested by 15 volunteers and managed to identify people correctly 93 per cent of the time.
The software does not use facial recognition systems to locate individuals since it is unlikely that the users will be looking straight into the Google Glass’ camera, InSight developer Srihari Nelakuditi told New Scientist.
He also noted that the ‘fashion fingerprint’ lasts only as long as the user does not change clothes. Afterwards, a new snapshot must be made and a new ‘fingerprint’ created. Thus, for privacy protection, all the user has to do is change outfits, argues Nelakuditi.
