LONDON: Google researchers have claimed that the new artificial intelligence system called ‘FaceNet’, is one of the most accurate ways of identifying human faces.
Three Google researchers, Florian Schroff, Dmitry Kalenichenko and James Philbin, said in research paper that their system achieved a new record accuracy of 99.63 per cent on the widely used Labelled Faces in the Wild (LFW) dataset. On YouTube Faces DB it achieved 95.12 per cent. “Our system cuts the error rate in comparison to the best published result by 30 per cent on both datasets,” the paper claims.
Facial recognition is a high-interest area for both companies and governments alike. In 2014, researchers from Facebook had published a paper claiming more than 97 per cent accuracy in recognising faces. Another group of Chinese researchers claimed better than 99 per cent accuracy, according to a report on Fortune.
These systems incorporate an artificial intelligence technique called deep learning and also include recognising voices and understanding the content of written text.
Aside from Google and Facebook, companies including Microsoft, Baidu, and Yahoo are also investing heavily in deep learning research, says the Fortune report.