Using deep learning algorithms, facial-recognition-technology (FRT) is becoming more and more effective.



Facial-recognition-technology (FRT) scans a person's face and then, using facial measurements, stores the data and allocates a code to that person.  FRT uses about 80 distinct nodal points on a person's face. This process, known as the biometric technique, looks at variables such as:

  • Shape of lips.
  • Type of eyes.
  • Length and width of the nose.
  • Depth of eye sockets

FRT can be used in the home or the workplace, using a two-step process::

  • Verification: comparing the person's claimed ID and face to the database
  • Identification: checking the person against all the faces in the database.

Using Deep learning, FRT is now able to identify people despite the following:

  • Variations in lighting.
  • Pose and expression variations.
  • Occlusion, sunglasses, masks, etc.
  • Appearance variation.
  • Pose and expression variations.

Worth considering the balance that needs to be maintained between privacy and security.