RealNetworks launches SAFR™ to help security professionals

RealNetworks launches SAFR

SAFR a facial recognition platform

RealNetworks, Inc. announced SAFR™ for Security, a new solution that integrates SAFR, the world’s premier facial recognition platform for live video, with leading Video Management Systems to provide enhanced visibility and situational awareness for security professionals.

Real-time face detection

The underlying SAFR platform has been optimized to detect and recognize faces in live video based on its industry-leading excellence in accuracy and performance. The SAFR platform can be deployed on a single PC to monitor a handful of IP cameras or scaled to thousands of cameras in a distributed architecture hosted on-premises, in the cloud, or hybrid. All data passed through SAFR is protected with AES 256 encryption in transit and at rest. The platform also provides actionable data for live analytics of traffic volumes, demographic composition, dwell times, and data exports for further reporting.

Available as a standalone solution or integrated with market-leading Video Management Systems, SAFR for Security provides vigilant 24/7 monitoring to detect and match millions of faces in real time, delivering a 99.86 percent accuracy rate. In the April 2019 National Institute of Standards and Technology (NIST) test results, the SAFR algorithm tested as both the fastest and most compact amongst algorithms for wild images with less than 0.025 FNMR. When SAFR for Security is paired with a VMS, the integrated experience includes video overlays within the VMS to identify strangers, threats, concerns, unrecognized persons, VIPs, employees, or other tagged individuals in live video. Security teams can customize real-time alerts to automatically notify them when persons of interest appear on a video camera feed, or make use of automated bookmarking to conduct forensic analysis. SAFR for Security attaches rich metadata to video footage so security professionals can search by time range, location, category, person type, and registered individual instead of sifting through hours of video to find a specific person.