The face recognition algorithm of Russia’s Vocord is the world’s most effective in its category, beating competitors including Google, according to a ranking produced by international online test platform MegaFace.
The MegaFace platform uses a database of more than 1 million photos depicting over 690,000 people photographed without posing and in natural lighting conditions.
Vocord’s algorithm is designed to recognize faces in large crowds of people such as at stations and airports. Photo credit: Wikimedia Commons.
Vocord, whose algorithm is inspired by deep neural networks, correctly identified 75.127% of the faces amid 1 million distractors, the best result in the competition.
“From the very beginning, we optimized our facial recognition algorithm for use in identifying a person in a crowd,” said Alexei Kadeishvili, Vocord’s technology director.
“That position was formulated as a result of the particulars of use of our system for searching for people in crowded places, such as stations, airports, malls and so on,” he explained.
Kadeishvili said the company is currently working on optimizing its face verification algorithms, which can be used in security procedures.
“We have some very interesting results related to counteracting fraud using facial authorization,” he added.
Vocord’s system is already in use in final products, and can process up to 3 million images for comparison per second.
Created in 1999, Vocord has made the news several times over the past few years with its innovative products. In addition to face recognition technologies, the company makes systems for recording traffic violations and monitoring traffic volumes and for use in video surveillance. Its software is used in more than 70 projects aimed at improving safety in cities, both in Russia and abroad.
The Russian company enjoys the support of Skolkovo, the international tech hub under completion on the outskirts of Moscow.
Vocord will present its Vocord FaceControl and Vocord FaceControl 3D products at the international Global Identity Summit later this month in Tampa, Florida.
