Making Face Recognition Safer

This post is also available in: עברית (Hebrew)

Recent progress in face recognition technologies also raises security concerns regarding privacy. Anticipating the use of such technologies by bad actors, companies are investing in the technological developments that would make face recognition safe. An existing example is an anti-spoofing technology, ensuring that the system is not fooled.
Face recognition technology searches an existing database of faces and compares them with the faces detected in the scene to find a match. It records faces from the camera and detects “people of interest,” providing real-time alerts upon detection, searches for similar faces from a single camera or across multiple cameras, detects any matches with faces on the watch list database and provide alerts, automatically create a log of people in the scene which can be used for forensic investigations, according to intelli-vision.com.
The new anti-spoofing technology ensures that system cannot be fooled by a photo or video image. IntelliVision, leader in Artificial Intelligence and deep learning video analytics software, is working on an anti-spoofing technology that will ensure that photos, screen-based images or even videos of a known face will not be recognized.
By offering different levels of liveness testing the anti-spoofing feature implements security without sacrificing convenience. The system can be configured for fast secure recognition, or highly secure recognition at a slightly slower speed.
Face recognition can be implemented on premise servers, over the cloud, or embedded in cameras with multi-core ARM chips such as those from Ambarella and Qualcomm. The face recognition system will soon be commercially available on NSC’s ELAN intelligent touch panels, adding a new level of intuitive convenience to the home control and automation platform.
“IntelliVision’s wide portfolio of video analytics have been used by OEMs and solution providers for many years,” according to company official. “Now, with a rapidly increasing number of leads from integrators needing to integrate video analytics to their projects, we felt it was time to put our existing services and support into a formal program.” This new Channel Partner Program will provide access to leads, volume discounts, demo software, marketing resources, training and support.
Face recognition is increasing in use not just for catching criminals and authenticating employees, but also for adding value through VIP identification in smart retail applications, and adding convenient personalization to the smart home.
The IntelliVision family of video analytics solutions includes object detection and classification (humans, vehicles, pets, airplanes), people/vehicle counting, demographics, intrusion/perimeter watch, object left/removed, license plate recognition (ALPR/ANPR), audio/sound recognition and facial recognition.