AiT network video recorders have many common types of analytics built in: line crossing detection; perimeter control; counting visitors with the ability to upload statistics; human detection; sound detection. Key features are new features: face detection and recognition; car recognition. The presence of second-generation face recognition technology distinguishes the NVR AiT from other models. They: provide higher recognition accuracy (up to 95%); allow you to detect more people in the frame (up to 64 people at a time); double the speed of recognition.If we classify the systems according to the level of responsibility for events, it will become clear that the video analytics introduced to respond to disturbing events is already quite perfect. So, if you introduce video analytics into fire safety systems, then you need to introduce another level, the so-called yellow, from false positives. Video analytics integrated into emergency services can work as the main system. But the decision is always made by a person, for example, a responsible duty officer, otherwise a collapse may occur. Opening a door instead of a concierge is quite realistic. Opening the barrier, for example, to enter the parking lot, is difficult because it is difficult to see and recognize the driver’s face. Reading numbers will protect unauthorized persons from entering the closed territory. Face control can be used: in access control systems, when, by the definition of a person, access to an object is possible (or not), which will exclude, for example, a concierge or security guard at the entrance; in crowded places (stadiums, variety venues, etc.) to identify wanted persons or to identify facial expressions of inappropriate people; in retail, hypermarkets, where you can use the customer base to determine the assortment of goods and recommend the goods at repeated visits. Passenger counting and statistical processing of the movement of people will be useful for developing the infrastructure of the city districts and the transport structure in order to optimize traffic. And if you combine all the data in the data center, it will help the city as a whole. The identification of numbers is advisable when searching for stolen vehicles, and in combination with access control means (automatic gates or barriers) simplifies access to parking lots and yards by access, reading the numbers.