Content provided by partner.
#FaceRecognition #Gesichtserkennung #AI #MachineLearning #VideoAnalytics #Sicherheit #Zugangskontrolle #Biometric #LiveVideo #FaceDetection

SAFR from RealNetworks: Live Video Facial Recognition

The powerful combination of accuracy, speed, and small size makes SAFR the best solution on the market for recognizing live video — where camera-unaware faces are in motion, at varying angles, in low lighting, at varying focal lengths, or partially obscured. The April 2019 National Institute of Standards and Technology (NIST) results rated SAFR as both the fastest and most compact among algorithms for wild images with a less than 0.022 False Non-Match Rate. With 99.86% accuracy, SAFR instantly detects and matches millions of faces in real time under challenging real-world conditions.

To unlock facial recognition’s full potential, systems must be scalable and adaptable. SAFR’s speed advantage is well suited to large-scale deployments: When it’s necessary to process hundreds of thousands of faces in real time, SAFR does so exponentially faster than any other solution on the market. Use cases include camera integration, live entertainment venues, public transit centers, retail insights, video monitoring, physical space analytics, and more. SAFR’s visionary approach is based on more than 20 years of world-renowned leadership in video technologies.

Enhanced Mask Detection & Recognition with Masks:

Available now, SAFR’s 3.0 release introduces a new default high sensitivity face detector. Customers running the high sensitivity face detector will see a 95.1% detection rate and 98.85% recognition accuracy rate for faces covered by PPE face masks — including non-surgical fabric masks of varying patterns — in surveillance-style videos of faces in motion. Detection efficiency has also been improved when multiple faces are simultaneously in the field of view to ensure detection and recognition speeds remain high.

SAFR 3.0 also includes a new mask detection algorithm which sets new standard for detecting all kind of masks and results can be visualized in a dashboard that enables to anonymously track mask usage rates and view and filter by age, gender, location, and time.

Occupancy Dashboard:

SAFR’s occupancy-counting dashboard helps businesses to automatically track compliance with limited number of people in enclosed spaces. SAFR can count the number of people entering and exiting a facility with a live dashboard that sends a notification when occupancy limits are reached. SAFR also tracks and reports on the percentage of individuals who are wearing masks, and includes age breakdowns in order to track sensitive groups such as seniors.

These solutions are made possible by the combination of high-precision MOBOTIX cameras and highly accurate technology from SAFR.

SAFR from RealNetworks

Key Features & Benefits

Live Analytics — Empower decision-making for scheduling, traffic flow, staffing, signage, merchandising, and more with an integrated understanding of pathing, gender, age, identity, and even sentiment.
Optimized for Live Video — The SAFR platform is optimized for accuracy and performance for live video. It can be deployed on a single computer with a handful of IP cameras or scaled to thousands of cameras in a distributed architecture.
SAFR Actions — Configure and manage responses to recognition events. Written in any scripting language, actions can be deployed for a wide range of IFTTT scenarios. Unlock a door, turn on a light, send an alert, record data for reporting, and more.
Speed & Accuracy: Delivers 99.86% accuracy in real time, detecting and matching faces in under 100ms. Has one of the lowest rates of bias for skin tone and gender, according to NIST, due to its massive and highly diverse global training set.
Flexibility: Deploys on premises or in the cloud. Runs on Windows, Mac, or Linux machines. Designed to connect, adapt, and extend to specialized applications, and custom integrations.
Reduced TCO: A distributed architecture means detection occurs at the edge while recognition occurs on the server, allowing for low-bandwidth environments. With the ability to leverage inexpensive GPUs, SAFR’s TCO is markedly lower than other platforms.
Privacy: Secures data with AES-256 encryption, in transit and at rest, and features full opt-in/opt-out capabilities. No data is passed over the internet if run on premises.
Deployment Neutral: SAFR can be deployed on premises or in the cloud, is OS independent, has a low memory footprint, is computationally efficient, and features multi-GPU support.

SAFR for Security

The highly adaptable SAFR platform is ideally suited for a range of real-world use cases where trusted, exceptionally accurate industrial-grade facial recognition is required. SAFR for Security integrates leading video management systems like Genetec, Milestone, and Digifort to provide security professionals with enhanced visibility and situational awareness in real time across any number of live video feeds. Whether on-prem or in the hybrid cloud — and whether Windows, Mac, or Linux — SAFR reduces the number of false alarms by recognizing security events as they happen, empowering security teams with the ability to respond immediately and better protect people, facilities, and valuable assets. When SAFR is paired with a VMS, the integrated experience can include: Video Overlays — SAFR features live video overlays within the VMS to identify strangers, threats, concerns, unrecognized persons, VIPs, employees, or other tagged individuals. Configurable Alerts — Security teams can customize real-time alerts and be instantly notified when persons of interest appear on a camera feed. Notifications can be further customized to initiate a wide range of powerful response actions. Automatic Bookmarks — Automatic bookmarks for conditional scenarios, with rich metadata attached, allow for more efficient investigative analysis with recorded video. Bookmarks are searchable by time range, location, category, person type, and registered individual. Live Analytics — SAFR provides actionable data for live analytics with rich metadata. View traffic volumes, demographic composition, dwell times, and data exports. Configure powerful custom actions and alarms based on recognition events, from turning on lights to initiating a building lockdown.

Focus Verticals & Use Cases

With a focus on education, healthcare, retail, live events, transportation, and infrastructure, SAFR facial recognition solutions that rely on MOBOTIX cameras are being deployed for numerous use cases, such as secure access, event and venue monitoring, locating lost children, school safety, camera integration, retail insights, physical space analytics, hospital infectious disease prevention, and more. With MOBOTIX leading the way in video surveillance cameras and SAFR’s position as the premier facial recognition platform for live video, this strategic partnership delivers a broad spectrum of innovative solutions. This year, a high-profile European museum deployed SAFR at its entrance to automatically count the number of people by time of day with demographic breakdowns, such as gender and age, regardless of how many faces are in the frame. The same system can also provide sentiment analysis that reflects museumgoers’ overall happiness or satisfaction. This solution is made possible through the combined excellence of versatile, high-precision MOBOTIX cameras and the highly accurate SAFR facial recognition platform.

Contact RealNetworks

Salutation *
I agree to the Data Protection Notice. *
I took note of my Right of Revocation. *
I agree to sending a copy to MOBOTIX AG. *