Oz Forensics, a Skolkovo resident company, has been ranked second in the world for its facial recognition technology, which underwent testing through Labeled Faces in the Wild (LFW) a public benchmark and dataset for facial verification. The technology in question, Oz Biometry, is an AI that can recognise faces with 99.87% accuracy within three seconds, placing it in the top tier.
The LFW test, created by the University of Massachusetts Amherst, is called “Unrestricted, Labeled Outside Data” and compares facial images that differ greatly in features such as lighting, age, weight, and so on. Following data protocol Unrestricted, Labeled Outside Data utilizes over 3 million images and 30k persons, so while it can accurately detect a face it also teaches computer vision.
Oz Forensics team. Photo: Oz Forensics
“We use different tools to develop and perfect the technology. The successful combination of all the latest developments allows for great results in accuracy in the world’s independent tests. The current measure recognizes our biometric technology, Oz Biometry, as a quality product. From the business point of view, having of a solid grade from an independent testing center will have a positive effect on a client’s decision in favour of one vendor over another,” said Artem Gerasimov, the CEO of Oz Forensics.
Where can Oz Biometry be applied in the real world? There are numerous areas, including the banking sector, security surveillance systems and the detection of so-called “deep fake” videos.
From a business standpoint, the banking sector is a key area for Oz Forensics as fraud prevention is an ongoing battle. Nearly all banking institutions put clients through rigorous security checks during processes such as remote client transactions, credit applications, customer authentication online and so on. The ability to identify a person quickly and accurately using Oz Biometry would increase both client conversion and security while reducing the bank’s expenditures on operators and SMS-mailing with OTP (one-time-password). According to Oz Forensics, this would also benefit bank customers as the time it takes to register would drop by 50% and credit card limits and online loans would double.
“The problem of remote biometric identification of users is more relevant now than ever before taking into account customers’ transition online and the internal business processes of companies,” said Mikhail Styugin, the head of the department of Information Security of the Skolkovo Foundation Information Technologies Cluster. “Oz Forensics has previously shown brilliant results in detecting artificially modified documents and now in recognizing biometric ‘deep fake’ attacks using deep learning neural networks. The expansion of the tech stack allows for the creation of new product solutions that are so important when taking into account the changes in the threat landscape in 2020.”
That is why having neural networks that can detect artificially modified documents has never been more important, both for the benefit of both businesses and customers.