Face picture high quality determination: An excellent threshold may be used through face conscription to make sure that just the finest quality face models are going to be stored in to database.

Ability to tolerate face position: Our face recognition program has a particular tolerance to face position which guarantees face enrolment expediency: rotator of a face can be as much as 10 degrees from front in every direction (nodded her head up or down, rotated left or right, moved left to right).

Many models of the same face: Biometric model report can include multiple face samples of the same individual. These examples can be signed up with different face postures as well as expressions, through different resources and in various time period therefore allowing to enhance matching high quality.

Identification capacity: Our system capabilities can be used within 1-to-1 matching (confirmation), as well as 1-to-many setting (identification).

Small face functions template: The face functions template uses up only 2.3 Kilobytes, therefore our programs can handle big face directories.

Features generalization setting: This setting generates the gathering of the generic face functions from several pictures of the same topic. Then, every face picture is refined, features tend to be extracted, and also the collections associated with features tend to be analyzed as well as combined right into a single generic features selection, which is created to the data source. This way, the actual enrolled function pattern is much more reliable and also the face recognition quality is raised considerably.