The growing use of biometric and authentication solutions, online and offline, has raised the risk of ‘spoofing’ attacks, an attempt to spoof the system with an artificial representation. Therefore, having robust technology to mitigate against spoofing is essential as part of a mix of tools to verify someone. This is true whether that be for verifying age, identity or authenticating a returning customer. The purpose of liveness is to make sure the person you are verifying is a real person. Liveness does not recognise who the person is (that’s facial recognition), and it does not check a face against faces in a database. It is most commonly used in combination with other authentication factors to ensure that authentication or verification isn’t being spoofed.
Yoti’s proprietary and patented solution. A liveness check uses a ‘deep neural network’ that determines if it is interfacing with a physically present human and not a bot, presentation attack (i.e. mask, screen or paper image) or injected data (i.e. recorded video images). A ‘deep neural network’ is an AI system that learns to recognise patterns by using many layers of interconnected "neurons" that work together to process information. In this case, the AI system has learnt to detect signs that it is not interfacing with a real person.
Please have a read of our whitepaper for more information.