One of the keys to understanding face recognition, it seems, is understanding how the brain comes to recognize voices.
For reliable face recognition, it is crucial that an acquired facial image is registered to a reference coordinate system.
In complex applications like face recognition, it is observed that one kind of feature set is not rich enough to capture the entire face information.
Automatic face recognition for still images with high quality can achieve satisfactory performance, but for video-based face recognition it is hard to attain similar levels of performance.
"It turns out that while the brain and the visual cortex is very good at many things, like face recognition, it is really really bad at analysing lighting and reflections".
(Facebook notes that people can opt out of the photo-tagging service by altering their privacy settings).Given the sensitivity, Google decided not to release a face-recognition search engine it had made.
Adobe has also improved its face-recognition technology to make it a bit faster to verify the correct association of names with faces.
In our conversation about Facebook's face-recognition software, he added, "It'll just create a more paranoid society with a fakey-fakey social life — much like what happened in Communist countries, where people had a fake social life that the Stasi could see, and then this underground life".
In the future, as face-recognition technology becomes more accurate, it will become even more intrusive, because of pressures to expand the biometric database.
Amit Roy-Chowdhury of the University of California, Riverside, in collaboration with Conrad Rudolph, an art historian, has applied face-recognition technology to see if it can solve such questions.
