MOSCOW, April 1. /TASS/. The smart cameras bought by Moscow authorities from the NtechLab start-up company are capable of recognizing faces partially hidden under masks or headgear, a spokesperson for the company told TASS on Wednesday.
"Our algorithm can recognize faces with a significant overlap of up to 40%. This includes faces partially covered with medical masks, head kerchiefs, motorcycle helmets and elaborate head gear as well as people turned to the camera in profile," the spokesperson said confirming that the Moscow authorities did purchase key components for the facial recognition network from NtechLab but, according to the latter, questions about that network’s use should be addressed to Moscow’s IT department.
The company explained that the NtechLab-developed systems are capable of analyzing video streams from unlimited number of cameras and the data on any number of people can be downloaded on it.
"As a rule, the network operates in the blacklist mode, which means that when a person added to the monitoring list appears in front of a camera, it identifies them and sends a corresponding notification," the spokesperson said. "The contractor, but not us, decide on who is on the list and where this notification goes to."
According to the interlocutor, big cities use similar networks to locate wanted criminals. The NtechLab FindFace Security system was used during the 2018 World Cup at the stadiums and fan zones to help identify more than 100 persons on the federal wanted list.
It was reported earlier that Moscow city lawmakers had put forward a proposal to use the current CCTV cameras for identifying violators of the mandatory coronavirus quarantine. The coronavirus-infected escapees from either home or hospital quarantine might be fined from 4,000 to 5,000 rubles (around $50 to $63). The chair of Moscow’s parliamentary commission on security, sport and youth policy, Kirill Shchitov, told TASS that the city’s CCTV network has more than 175,000 cameras. Their footage is also used for investigating about 70% of offences.
NtechLab is a Russian startup founded by computer programmer Artem Kukharenko and businessman Alexander Kabakov in 2015. It specializes in advanced methods for a self-learning neural network to create facial recognition algorithms.