Face Recognition can
be used in 1-to-1 matching (verification), as well as 1-to-many mode
(identification).
Identification (one-to-many searching):
To determine someone's identity in identification mode, Face Recognition quickly
computes the degree of overlap between the live face and those associated with
known individuals stored in a database of facial templates. It can return a
list of possible individuals ordered in diminishing score (yielding resembling
faces), or it can simply return the identity of the subject (the top match) and
an associated matching level.
Verification
(one-to-one matching): In verification mode, the face can be stored on a smart
card or in a computerized record. The system simply matches the live face to
the stored one. If the matching score exceeds a certain threshold, then the
match is successful and identity is verified.
What makes the Face Recognition powerful is the fact that it is resistant to changes
in skin tone, facial hair, hair style, eyeglasses, expression and poses and
hence depends on the intrinsic shape and features of the face. In addition, it
has been found that the live face contains enough information to accurately
distinguish an individual amongst millions of people.
