Identifying people by their faces is the most natural thing in the world, its human instinct. The problems arise when we let the machines take over. SmilePass brings the human element back into secure biometrics.
The SmilePass Biometric Engine, uses complex matching algorithms to replicate, and improve on the job done by our brains. It works by measuring facial features, and aspects that are only identifiable, on a live human face.
The engine processes these facial images and creates digital ‘templates’, which are stored for future comparison. These templates alone carry no personal data and cannot be recalculated back into a picture – Therefore it is impossible to compromise an identity.
Facial recognition is typically used in one-to-one (1:1) identification, i.e.) matching a live face against a record of the person it is supposed to be. This method would require an initial identifier, such as a password or username, to recall the relevant person’s template for matching.
A one-to-many (1:n) scenario is commonly used when you want to identify one person against a database of people. It is possible to set acceptance thresholds to whatever level required for purpose e.g. you may want to return a list of possible matches, a sort of ‘rogue’s gallery, for further human identification.
The applications of facial recognition are innumerable, depending on your requirements. Try out or demos below to get a feel for how the technology can be applied
Face recognition can be used in isolation or with SmilePass Voice Recognition for increased, multi-modal, security.
Facial Detection Points