The technology, which is limited to high-resolution photos, showed promise in a series of tests where participants were asked to identify people from these images.
Those who were unfamiliar with the subject's face were able to successfully match them with another photo of that subject 71 percent of the time, while those who were familiar with the subject's face had an accuracy rate of 84 percent. In a second test, observers were able to reliably name a familiar face from an eye reflection image.
"Corneal reflections can reveal surprisingly rich information about the social environment," wrote Rob Jenkins and Christie Kerr in their paper, published in PLOS One. "These bystander images were not merely informative about facial appearance, they were properly identifiable to viewers who knew the faces."
While the researchers used very high-resolution cameras in their tests, they note that "pixel count per dollar for digital cameras has been doubling approximately every twelve months. This trajectory implies that mobile phones could soon carry >39 megapixel cameras routinely."
They added: "For crimes in which the victims are photographed (e.g., hostage taking, child sex abuse), reflections in the eyes of the photographic subject could help to identify perpetrators."
0 comments:
Post a Comment