Google glass: App measures facial expression

Technology News |
By eeNews Europe

Besides interpreting the facial expression, the Google Glass app ‘Glassware’ also provides an estimate of the person’s age, sex and other personal features. For reasons of data privacy, the algorithm is not designed to identify a person. All computations are performed in the Google Glass CPU in real-time; there is no need to transfer the image data for fast processing to another location. The Fraunhofer app for Google Glass is based on an existing facial recognition and analysis software library named SHORE. It has now been shown for the first time running on Google’s device.

According to the Fraunhofer researchers, the software opens up a host of application options for data glasses. Examples could be auxiliary tools for persons suffering from autistic conditions: Many autistic persons have problems to correctly identify the emotional content of a facial expression. Data glasses could superimpose this information into the person’s range of vision. Likewise, visually impaired persons could benefit from the software: The information about a person’s facial expression could be converted into an audible signal and fed into an earphone, for instance. Conceivable are also interactive games or market research applications.

SHORE is a C++ library that has been compiled over many years of research activity. It enables users to detect faces and other objects in high detail. The high optimisation of the software makes it possible to adapt it for a broad range of hardware platforms and operating systems, in particular for mobile devices.


Linked Articles
eeNews Europe