A biometric template is a stored representation of biometric data. But it is not the data itself – and this is the crucial distinction between biometric data and a biometric template.
In order to create a biometric template, first the source data must be collected by a sensor. So for example, this could be a person’s facial image, a voice recording, or a fingerprint scan – recorded by a camera, microphone, or fingerprint pad respectively.
The capture of a biometric template usually happens during the onboarding process, as this is the first time a user interacts with a biometric system (i.e. when opening a new account with a bank).
The captured biometric data is then analyzed through various algorithms and mathematical models in order to convert that data into a biometric template. This data becomes a master profile from which the unique features of the individual’s face, hand, finger, iris, or voice are extracted, analyzed and then converted into a mathematical file. Depending on the vendor, biometric templates may be pseudonymized.
iProov complies with GDPR, which means that biometric data can only be used under agreed and specific terms (user authentication, fraud detection, and maintenance of these methods). iProov is exhaustively tested regularly by governments and enterprises. Learn more about our governance here.
It’s important to remember that the biometric template is not the same as the original data captured. It is a representation of that data in a different form. This generally means that stealing a biometric template would be useless for a fraudster. In iProov’s case, instead of imagery the fraudster would find an anonymized binary code. Ultimately the template functions as a unique representation of the person, but it is not an image.
The exact mechanism of a biometric template and the privacy rules around them will vary from company to company. Read more about finding the right biometric vendor here.