June 22, 2017 -
FindFace is finding applications in public safety, law enforcement and fraud prevention, according to a report by PYMNTS.com.
Developed by Russian machine learning and artificial intelligence company NTechLab two years ago, the technology initially began as a tool for identifying people on Russian social media from pictures of faces taken with a smartphone camera.
To address privacy concerns regarding the use of biometrics, NTechLab CEO Mikhail Ivanov said that society needs to maintain an ongoing dialogue about when and where to apply facial recognition technology.
“We have to regulate how to use it in the future. It already exists,” Ivanov said. “Everything can be useful for good things and for bad things. I think there are lots of cases where facial recognition can be used for good things to make our lives safer and more comfortable.”
Ivanov said the government already has personal data about its citizens, and that biometrics and facial recognition are simply another piece of information it can use to help protect people in a public safety and law enforcement capacity.
FindFace can identify and detect criminals based on biometrics data stored on the police database, an application that has already been tapped by several law enforcement agencies across the world.
NTechLab is hoping that more retailers will adopt FindFace technology to implement targeted ads and promotions based on consumer buying habits, loyalty rewards programs, and preventing theft and fraud.
FindFace can prevent fraud in financial services by requiring clients to enter their biometrics in order to access their personal data and accounts, while car insurers can use the technology to confirm that the authorized driver was actually the same person driving the vehicle at a particular time and place.
Using deep learning strategies, NTechLabs engineers are able to teach the FindFace neural network to recognize the differences between faces.
In the first stage, engineers show photos to the artificial intelligence system and ask it to identify which images contain faces.
They correct errors and add more complex layers until the system is able to distinguish between a whole range of races including Asian, European, African. The system then learns to look beyond any obstructions such as hats, glasses or face masks.
FindFace currently has the ability to recognize faces with 73.3 percent accuracy, which is 5 percent higher than Google’s facial recognition technology.
However, the lack of understanding of how the technology can be implemented or how it can help businesses, along with personal privacy concerns, have prevented the technology from gaining significant traction.
NTechLab has been approaching various companies to integrate FindFace into their business, and not the other way around, Ivanov said.
To date, the technology has been licensed by less than 100 companies, however, there are several pilots currently underway.
Ivanov expects to deploy FindFace in several markets including Russia, Europe, the U.S. and Asia in the near future.
Last October, NTechlab released FindFace.Pro, a facial recognition as a service for businesses.