Face recognition software, like may things in the technology realm, continues to get better. It is to the point that now, according to the results of a recent contest, that machines can outperform most humans at recognizing faces. The Face Recognition Vendor Test (FRVT) 2006 and the Iris Challenge Evaluation (ICE) 2006 showed that the ability of computers to recognize faces has increased ten fold since 2002, and one hundred fold since 1995. The results are due mainly to the increase in image resolution and better recognition algorithms according to Jonathon Phillips. Jonathon Phillips is the program manager for the NIST tests and lead author of the agency’s report.

A part of the success of the software is the ability to look at certain features of the face that do not change over time. This includes the curves of the eye sockets, nose, and chin. This has the benefit of better recognition in adverse lighting conditions. The software can better recognize the shapes of these areas far better in bad lighting, as they stand out better.

Source: Technology Review