Facebook working on 'almost human' tech to recognize faces
Facebook is reportedly working on a facial recognition technology that is touted to be nearing human-level precision.
The project, dubbed DeepFace, is aimed at verifying whether two people in side-by-side photos are the same.
Facebook said that the technology provides 97.25 percent accurate results, which is just shy the 97.5 percent humans have scored in the same standardized test, The Verge reports.
According to the report, the technology maps out 3D facial features and then makes a flat model that's filtered by color to characterize specific facial elements to match the given images.
The social networking giant said that it has tapped into a pool of 4.4 million labeled faces from 4,030 different people on its network in order to help the system learn.
The report said that Facebook had introduced facial recognition feature, the actual capability to figure out who a person is in a photo, in late-2010 for US users before a worldwide rollout in 2011.
Facebook, which gets more than 350 million photos uploaded by users each day, has pointed to the tool as a way to speed up tagging, the report added.
(Posted on 20-03-2014)