NtechLab: AI driven video analytics based on face, silhouette and vehicle recognition

As the industry of video surveillance grows so do new security technologies that allow us to easily monitor and secure private or important areas. Recent improvements in video analytics have been a game-changer, ranging from applications that count people at events, to automatic license plate recognition, along with other more well-known scenarios such as facial recognition or smart parking.

NtechLab is an accomplished provider of video analytics solutions powered by artificial intelligence. The organization’s chief product FindFace Multi provides record-breaking accuracy and high speed with face, body, and human action recognition as well as vehicle detection.

The story behind NtechLab

Artem Kukharenko is the man behind NtechLab technology. He has specialized in neural networks since his time studying at Moscow State University where he authored several scientific works on the subject. In 2015, he developed a smartphone app that could specify a dog’s breed from a single photo: called Magic Dog. Through a friend he was introduced to businessman and passionate tech-investor Alexander Kabakov (who has also funded popular Russian online service YouDo, helping users to find people to help in solving everyday tasks and household chores). Kabakov suggested a pivotal change in the use of the technology behind Magic Dog; creating an algorithm that could instead recognize human faces. Kabakov and his partners became first-round investors in NtechLab, and six months later the company’s product had beaten Google in the MegaFace Challenge, which pits facial recognition technologies head-to-head, organized by the University of Washington.

In conversation with Andrei Telenkov, Chief Executive Officer

Q. Trust is one of the most important currencies of leadership that requires authenticity and consistency to maintain. What’s your take on this?

As a leading facial recognition software company, we have established a worldwide network of trusted partners who implement our technology. Our algorithm works with unique facial features which makes it impossible to restore the original image of the face; something that is of distinct importance in ensuring personal data protection.

For commercial applications our software collects anonymised data, limited to relative age and gender in order to generate consumer data and insight that enables businesses to improve brand communication and customer journeys. In cases where a business is required to react to certain people (for example, the VIP-scenario, where businesses use the system to recognize special guests to offer them VIP services) we expect businesses to receive written confirmation from individuals authorizing the use of their images for such purposes.

Q. What are the different products NtechLab offers? And what can you tell us about their (products) association with FindFace?

NtechLab’s flagship product is FindFace Multi, a groundbreaking tool that accurately detects faces and silhouettes of people, and cars in real-time video streams.

Using an intelligent criteria-based facial search, the system can detect and track ‘persons of interest’ in a split second. It also has high levels of accuracy in detecting faces when they are partially covered with medical mask or clothes. Silhouette recognition enables the identification of people by the color of clothes and accessories. Vehicle recognition identifies the type of body, color, manufacturer, and model of a car. It works even if license plates, or part of the car, are not visible.

Q. In what industries are your clients? Can you provide us with one or two success stories describing the challenges your clients faced and how your solutions helped?

NtechLab algorithms are designed to help organizations and businesses make better decisions. The technology empowers governments and security services to act with greater speed and confidence and provides commercial companies with richer audience analytics. Founded in 2015, NtechLab now is a world-renowned expert in video analytics with more than 150,000 cameras in 20 countries globally connected the company’s face recognition software.

As an example, I would like to cite a major agreement between FESCO Transport Group and NtechLab, that will create landmark ‘smart’ shipping port at the Vladivostok Commercial Sea Port in Russia. NtechLab is providing a unified video analytics system to monitor labor and transport safety, based on a common biometric database created by the sea port. The technology delivers remote systems access and the ability to monitor the port’s operational processes in real time. One aspect of NtechLab’s systems is the ability to manage the access control to the enterprises with far greater efficiency by using biometrics. Our platform will also be able to identify vehicles, and to read container numbers during loading and unloading to compare and confirm the accuracy of shipping and cargo information.

Q. NtechLab and Moscow Department of Information Technology work hand in hand, providing video analytics in Moscow city. That said, how intelligent are the company’s face recognition solutions?

NtechLab’s algorithm is integrated into the Moscow city video analytics system. The functionality of the solution allows real-time high-precision face recognition in the video stream, comparing the results with databases of wanted persons. If a match is found, the system instantly sends alerts to law enforcement officers. The whole process from the actual appearance of a person in front of the camera to receiving a signal takes less than a second, which allows to quickly respond to any situation. The solution can be also used to find missing people. In 2020, over 5,000 crimes were solved in Moscow using a video analytics system with face recognition, according to the data of the Ministry of Internal Affairs of Russia.

Q. Please walk us through the awards and recognitions NtechLab has received so far?

In October, NtechLab wins first G20 innovation award for the best artificial intelligence startup. The event involved 22 countries, 100 startups, over 100 VC & Corporates. Among the 10 winners some revolutionary innovations in the Cleantech, Artificial Intelligence, Sustainable Mobility, Health and Internet of Things. A special event organized in the framework of the Italian Presidency of the G20, which brought together 100 startups from the G20 countries, more than one hundred venture capitalists, and big international companies. At the first stage of the competition, the most promising startups were nominated by each of the G20 countries. At the second stage, 100 startups that made it to the finals were evaluated by venture capital investors based on financial results, product uniqueness, team competencies, innovative ideas to tackle global challenges, the ability to build a more inclusive and sustainable future and other criteria.

In 2021, the company topped the Face Recognition Vendor Test (FRVT) undertaken by the National Institute of Standards and Technology (NIST) of the US Department of Commerce. Based on the results of seven independent tests, the algorithm developed by NtechLab was given worldwide top ranking.

Q. About the future, where do you see NtechLab a couple of years from now?

When creating the company, we chose between what we were used to and what we dreamed of – making cities better and safer. We observe that our customers have begun to look at video analytics functions more broadly. Our concept helps people with disabilities, children, and mothers with strollers. City cameras with video analytics systems can collect data, which, after analysis, make it possible to understand that a certain section of the road, for example, a crossing, is very often used by children and there is a threat to their lives due to the high traffic flow.

Meet the luminary

Before being named CEO in March 2020, Andrei was managing director at Vardanyan, Broitman and Partners (VB Partners), an investment company and one of the NtechLab shareholders.

Andrei represented the NtechLab at the first G20 Innovation League, a special event organized in the framework of the Italian Presidency of the G20, where the company received first award for the best artificial intelligence startup. Under his leadership, NtechLab also regained top positions in the NIST Face Recognition Vendor Test (FRVT), the most respected benchmark when it comes to evaluation of facial recognition algorithms.