NtechLab, a technology partner of Russia’s Rostec Corporation, won the Face Recognition Vendor Test (FRVT) competition of the U.S. National Institute of Standards and Technology of the US Department of Commerce (NIST).
The NtechLab algorithm was recognized by the American institute as the best in the world based on the results of seven independent tests, and in three of them a record was set in the entire history of testing, Rostec announced.
NIST's FRVT Testing is the only globally recognized face recognition competition; including helping the US Department of Commerce identify the world's best providers of such software solutions. It corresponds to the scenario of confirming a person's identity from a photograph.
This scenario is used in a wide range of civil, law enforcement and national security programs, including photo verification on visa documents and when issuing passports, as well as in payment systems using biometric data (face-to-face).
The FRVT evaluated over 100 algorithms from developers from a wide range of countries, including China, the United States, and Israel. FRVT testing was carried out on several photo bases. When summarizing the results, the accuracy and speed of the search were taken into account, as well as the adaptability of the algorithm to subsequent changes.
“In six years, NtechLab has grown from a promising startup into a company with more than a hundred employees, thanks to which it attracted Rostec's attention at the early stages of its development and has gone on to become one of the leading companies in the Russian IT market,” said Vasily Brovko, Chairman of the Board of Directors of NtechLab and Director for Special Assignments of State Corporation Rostec.
NtechLab algorithms have repeatedly proven their technical superiority in representative international competitions. In 2018, NtechLab became one of the three winners of the WIDER Pedestrian Challenge competition for detecting pedestrians based on their silhouettes, in 2019 it took second place in the ActEV-PC international competition for recognizing human actions in video.