Artificial intelligence has always been an important component of biometric technology. In a sense, biometric recognition is artificial intelligence, enabling a computer system to correctly match a fingerprint or an iris pattern to a template on file. But it’s undeniable that AI in its broader sense is coming to play a bigger role in biometric technologies. It is no longer sufficient for a face-based authentication system to match two static images, for example; now, it needs an AI boost to check for signs of liveness. Meanwhile, fingerprint scanning systems are increasingly expected to adapt their algorithms to recognize an end user’s biometrics more reliably, and more quickly.
With these advances, the line between AI and biometrics is blurrier than ever. Machine learning and neural networks have evolved to deliver rudimentary ‘computer vision’ that can identify things like weapons or suspicious behaviors in much the same way that biometric facial recognition and iris recognition can provide matches between two different images, and we’re now seeing a melding of these fields of specialization across a range of sectors and applications.
Here are a few of the latest examples of how this major trend is unfolding:
AI-driven facial recognition and computer vision are playing a growing role in smartphones, allowing them to not only identify their users but other objects and settings at which their cameras are pointed:
Meanwhile, computer vision is being paired with more traditional facial recognition to find new and expanded application areas:
Investors appear to be increasingly excited about newer challengers developing biometric and computer vision technology:
… and more well-established biometrics specialists are amping up their AI R&D: