SAFR algorithm performance metrics are provided below. They reflect achieved accuracy using the LFW (Labeled Faces in the Wild) dataset in the “unconstraine...

Match probability reporting allows operators to distinguish between exact matches and probable matches, enabling quicker responses to emerging situations. T...

SAFR can reliably match against millions of faces in under a second.

Yes.  You can easily monitor such information by using SAFR's Occupancy Dashboard, Occupancy Alarm Dashboard, or Occupancy Report features. 

SAFR performs non-literal matching which handles multiple variations, including masks, makeup, spectacles, hair styles, and large variations of age for the ...

This functionality isn't native to SAFR. However, a 3rd party system integrator could collect this data and store it in the face's person record.

Yes. SAFR allows all personally identifiable information (PII) to be either not be stored or, if stored, purged either automatically through lifecycle polic...

This functionality isn't currently supported.

Our system works well as long as a face picture has at least 2 out of: eyes, nose, and mouth. You can configure the identity matching threshold so that SAFR...

It's highly unlikely that a forensic sketch will ever result in a positive face recognition match with any facial recognition system. Setches are never ...