Computer vision to enhance security and user experience

SAFR from RealNetworks offers accurate, fast, unbiased face recognition and additional computer vision features. Optimized to run on virtually any camera — or camera-enabled device with our SDKs — find out how you can leverage the power of AI to enhance security and convenience.

 

Face recognition for masked faces

 

Real-time automated surveillance

 

Contactless, secure check in

Why SAFR?

The fastest, lightest and most accurate algorithms for live video.

  • Reduces the total cost of ownership
  • Ideal for IOT ready edge devices
  • Real time analysis for actionable intelligence
Mask Detection

Fast
< 100 ms recognition speeds

Recognition of Masked Faces

Accurate
99.87% LFW recognition accuracy, 98.85% for masked faces

Unbiased

Unbiased
Least biased, globally available recognition algorithm

Occupancy Counting

Secure
Built for privacy with AES-256 encryption and offline deployment options

Contactless Secure Access

US Company
Headquartered in Seattle, with a global network of sales and support

LATEST FROM SAFR

SAFR® from RealNetworks Awarded its Third Small Business Innovation Research (SBIR) Contract from the United States Air Force (USAF)

Industry Leading SAFR® Facial Recognition for Live Video Integrated with Geutebrück VMS

SAFR® from RealNetworks Launches Version 3.4 Featuring Passive Liveness Detection & Anti-Spoofing

Release Notes: Enhanced Liveness and SMS Watchlists for accurate, secure and customer-friendly access control workflow

What we do best

 

Touchless Access Control

Face-based access control is more secure, and more user friendly, than physical authentication methods.

Integrate SAFR with existing secure access technology to allow registered individuals to unlock a door using just their face — or enable dual factor authentication. Learn more ›

 

Automated Surveillance

Manual monitoring is expensive and inefficient. AI can perform real-time, automated identification of persons of interest and prevent security incidents, freeing up your security team to do more.

Integrate with a VMS or deploy standalone as your security monitoring solution — on one, or thousands of cameras. Learn more ›

Identity Authentication

Identity Authentication

Leverage SAFR SDKs and Web APIs to enable face-based identity authentication for remote and in-person workflows.

Ensure maximum security for bring your own device (BYOD) applications such as remote testing user verification. Learn more ›

With automated real-time alerts, more efficient post-event analysis, and ongoing actionable insights, SAFR frees up security and operations teams to do more.

COVID-19 response

As the world responds to the COVID-19 pandemic and more people wear masks in public, computer vision platforms need to learn to recognize new things.

SAFR has a 95% mask detection rate and recognizes masked faces with 98.85% accuracy — ensuring no disruption of security workflows and unlocking new COVID-19 prevention use cases.

Learn more ›

Masked Recognition

Get started

SAFR brings high-performance computer vision to virtually any camera. Deploy SAFR’s core features as a standalone solution via the SAFR platform, via our SDKs and web APIs, integrated with many leading video management systems, or out of the box on a camera already running SAFR such as the AXIS P3255.

Recognition

Face Recognition

Occupancy Counting

Person Detection

Occupancy Counting

Occupancy Counting

Fake faces

Liveness & Anti-Spoofing

Demographic Characterization

Demographic Characterization

Real-time Alerts

Real-Time Alerts

Event Generation & Reporting

Event Generation & Reporting

Recognition of Masked Faces

Recognition with Masks

Contact us to request a free demo or speak with a technical representative about our full feature list.

Partners

We work with leading technology partners across the globe. Look for the AI powered by SAFR logo. Learn more ›
Axis and SAFR

Contact Us

Request a free demo or get in contact with one of our sales or technical representatives to learn more.

  • This field is for validation purposes and should be left unchanged.