In a groundbreaking move NtechLab, one of the world leaders in video analytics, has released FindFace Multi, a detection technology that uses an advanced algorithm to recognize not only faces, but also silhouettes of people and cars.
This update to its flagship product is a step change for security experts in their fight against crime and also an invaluable tool for commercial planners. Another distinctive feature is its ability to support many video streams and facial database entries, ensuring the system offers unrivalled scalability.
Andrei Telenkov, CEO, NtechLab, said: “The solution is of serious interest to our clients and potential customers, because practically none of our competitors has a similar platform for multi-object video analytics, when cameras can be configured for different types of objects, and at the same time one camera can process several different types of objects at once,”
Silhouette recognition allows FindFace Multi users to count and search, as well as track movements, with high accuracy the numbers of people moving through an environment as well as identifying individuals. The algorithm also takes into account markers such as height, color of clothes and accessories. The ability to accurately identify numbers is also an important feature for urban planning, shopping center developers and managers and other commercial areas.
The vehicle recognition function determines the body type, color, manufacturer, and model of a car, as well as searching by license plate. Even if license plates, or parts of the vehicle are not visible or obscured, the system can still identify a car.
Using an intelligent criteria-based facial search, a user can detect and track ‘persons of interest’ in a split second. It also has high levels of accuracy in detecting faces even when they are partially covered with medical mask or clothes. The algorithm was recently recognized in a Face Recognition Vendor Test competition, run by the National Institute of Standards and Technology, US Department of Commerce.
The algorithm took first place in the global competition because it increased, by a factor of ten, the ability to accurately identify faces that are masked, as well as doubling benchmark accuracy rates of individuals who weren’t wearing masks. It can, for instance, accurately identify someone whether they have a beard or are wearing glasses or a mask.
Liana Meliksetyan, International Commercial Director, NtechLab is anticipating significant growth for FindFace Multi: “We have launched pilot projects in Russia and other CIS countries, Latin America and the Gulf. By the end of the year, we plan to implement more than 20 large projects in such industries as urban video analytics, transport management, construction and retail.”