The Guardian reports from a paper in Science that researchers from the University of Glasgow have developed software that increases the efficacy of facial recognition software to one hundred percent. The experiment was based on comparing faces with celebrities....
Jenkins and his colleague Professor Mike Burton tested the averaging approach using FaceVACS, a system that is being tested at Sydney airport.
The website MyHeritage.com uses the software in a celebrity lookalike service. Surfers submit images of themselves to the site, which matches them to the nearest celebrity picture on its database of more than 31,000 photographs. Jenkins and Burton submitted images of 459 celebrities they knew were on the database. The system matched them to the correct celebrity 54% of the time. When the pair created average celebrity faces from 10 images and resubmitted them to the website, the software was correct 100% of the time. "That's the first time anyone's reported anything like that level of accuracy on such a variable set of images," said Jenkins.
To make the test even tougher, they created averages using only those images that the software had been unable to match in the first test. In this test the software recognised the averaged faces correctly 80% of the time. "We have fixed the baseline at zero, so any improvement we can attribute just to the averaging process," he said.
This represents an important escalation in the efficiency of facial software and the capability of the authorities to track individuals in a surveillance society such as the United Kingdom. It does not require any upgrades to surveillance infrastructure. The improvement in software also serves as a reminder that restraints upon surveillance depend upon the laws governing the use of such cameras. The current incursions into our privacy will be worsened by the enthusiasm of government.