SAFR Standard: Customer Hosted (Perpetual)

Face recognition

SAFR is an exceptionally accurate AI-powered facial recognition system with a high degree of accuracy. SAFR uses the video images from an (Axis) IP camera to perform this analysis and to control peripherals such as cameras, door locks or alarm systems very quickly. With this software, access is managed based on people's identities.

Compatibility

SAFR works on a variety of operating systems, including Windows, macOS, Linux, iOS and Android.

To use this software, an on-site server is required on which the video images are analyzed. The recognition is carried out in real time and can be monitored immediately. The detection of mouth masks at incoming persons can also be immediately detected and linked to an alarm when no mouth mask is present.

AXIS Q1615-MKIII can run the SARF inside ACAP, allowing some of the analysis to be done on-camera. This saves server capacity and the analysis can be carried out even faster.

Functionality

Specifically, SAFR Inside's video feeds can do the following:

• Detect faces
• Detect when people are wearing masks
• Identify a person's age
• Identify the gender of a person
• Identify a person's sentiment (i.e. happy or unhappy)
• Learn and recognize faces, regardless of whether they are wearing a mask or not
• Learn and recognize faces that are partially closed off by sunglasses, caps, scarves, etc.