Of all the biometric modalities, facial recognition has become one of the most widely used and discussed in recent years. That’s why November is Face Biometrics Month at FindBiometrics, in which we’ll be delivering in-depth articles and other special features dedicated to this technology.
It’s a broad category though, with facial recognition deployed across a range of different applications – from consumer-facing finance, to frictionless enterprise security, to law enforcement – sometimes to the joy and acclaim of consumers, and sometimes to cries of overreach from privacy advocates. So the best place to start is probably to take a look at some of the most prominent kinds of facial recognition technologies and applications in order to understand the big picture regarding this high-profile modality.
Let’s start with one of the most universally acclaimed uses of facial recognition – smartphone authentication. With most smartphones featuring built-in cameras for several years now, the hardware has long been in place to support mobile face scanning. And indeed, such systems have been in the marketplace for quite a while, with mobile biometrics pioneer Daon supporting facial recognition through its IdentityX platform since at least 2015. But it was Apple’s switch from fingerprint recognition to 3D face scanning in its latest smartphones – starting with 2017’s iPhone X – that really prompted the mobile industry more broadly to support face-scanning systems that could quickly and conveniently unlock users’ smartphones. Few of these are as sophisticated as Apple’s Face ID system, which is based on three-dimensional imaging, but the company’s rivals are racing to catch up, with Samsung and Google both focusing on facial recognition in their newest smartphones as well.
Tied to this trend is the rise of selfie-based onboarding and authentication. This variation on the smartphone ‘face unlock’ concept combines mobile facial recognition with AI-driven document reading in order to verify user identity: essentially, the user takes a selfie, and also uses their phone to capture images of official identity documents like a driver’s license, and then the system can match the user’s selfie to the image on the ID. Some of these systems, such as those offered by Onfido, FacePhi, and Jumio, are so accurate that they can actually enable banks to let customers remotely open new accounts, and indeed the financial services sector is the area where selfie onboarding has really taken off in recent years. But these systems can also be used as the foundation for standalone digital identity programs, as can be seen in the efforts of Yoti and its mobile ID platform.
Biometric Border Screening Takes Off
Moving from the consumer-facing side of things to the government side, facial recognition has also seen pronounced growth in recent years in the realm of biometric border control, particularly with respect to airports in the US. Face scanning is increasingly being used to match travelers against their passports and other ID documents; and while these efforts are being spearheaded by government agencies like US Customs and Border Protection, they’re also being embraced by a growing number of private sector partners in the travel industry, with airport and airline administrators recognizing the efficiency benefits of biometrically scanning passengers. Despite this enthusiasm, and some early indications that travelers also appreciate the convenience of this technology, deployments have raised alarm among privacy and civil liberties groups like the ACLU, which has decried the rapid expansion of the CBP’s face scanning programs in particular.
This brings us to the most controversial aspect of facial recognition technology – its use in law enforcement and surveillance. A growing number of security and police agencies have embraced the use of facial recognition to track down criminals, and even to surveil public areas. But this became an explosive public issue after the ACLU publicized Amazon’s sale of facial recognition technology to law enforcement agencies in the spring of 2018. The reporting prompted a serious public backlash against such practices – possibly due to an already negative perception of Amazon over some of its other business practices – and has now led to multiple legislative attempts to limit the government’s use of facial recognition across the country. Police agencies tend to insist that facial recognition technology offers them a useful tool in the fight against crime; and broadly speaking, their use of the technology in the US and the UK pales in comparison to the intensive biometric surveillance programs taking root in China. But the discourse over this issue is ongoing, with 2020 presidential race hopefuls like Bernie Sanders making the regulation of facial recognition a policy plank for their campaigns, and some industry leaders, such as Microsoft and computer vision specialist Paravision, openly welcoming a balanced approach to regulation.
There’s a lot more to say about this issue, and much of it will come up in future Face Biometrics Month coverage. For now, though, what’s clear is that the government’s use of facial recognition technology has become extremely prominent even as consumers in the mobile and financial services markets have enjoyed the benefits of face biometrics technology in an entirely different way. There is great breadth to how this technology can and should be used, and that’s why it’s one of the most important modalities in the biometrics industry today.
Stay tuned to FindBiometrics as we bring you more in-depth coverage and analysis through the course of Face Biometrics Month.
November 7, 2019 – by Alex Perala