Russian programmers have repeated the success of their colleagues at the global competition for face recognition.
The Russian “Vocord” company issued a press release announcing that it surpassed all its competitors in face recognition technology, conducting a test of their algorithm on the database of the global Megaface Challenge competition hold by the University of Washington.
Thus, local programmers have repeated the success of their Russian colleagues from the N-Tech.Lab company that won this competition back in September of last year. The Megaface competition started at the Washington University in September of last year. Nearly 100 teams from all around the world had only 20 days to adapt their algorithms to the tests on a unique base of one million photographs.
As a result of testing, the first place was taken by the programmers from the Russian N-Tech.Lab company, their algorithm showed 73.3 percent accuracy, surpassing the teams of Google, Beijing University, and many others. After the competition was finished, the organizers decided not to close the database of photos and give everyone an opportunity to submit their own algorithm to determine the accuracy. As “Lenta.ru” found out, these conditions allowed the programmers from “Vocord” to repeat the success of their Russian colleagues, showing the accuracy of 75.127 percent.
Author: Anna Dorozhkina