|
There's something wrong with translating an article about something occurring in my own country, but hey...Yomiuri is a large newspaper, so there's no reason why they couldn't find out about something before other American sources. Thus the reason I'm posting this article - I don't see it anywhere else yet. Not a direct translation (yet), but I promise my article will contain the same information. Anyway, it's about being able to tell what someone is looking at by recording a high resolution image of their eyeball.
A team at the University of Columbia has developed technology to use high resolution digital cameras to record what is being seen by a person's eyes. (No, this doesn't work by embedding cameras inside eyes, go away Shadowrun.) Instead, it works more along the lines of things like being able to tell what a person in an old photograph is looking at, or being able to extract important information such as where terrorists are based on the reflections (in their eyes) of declarations they are reading.
Professor Shree Nayar and Dr. Ko Nishino of Columbia University intricately analyzed the relationship between the reflections on the cornea of the eyeball and the image actually perceived by the retina after passing through the pupil. The curved reflection on the cornea is recorded with a 6MP camera, and is then restored to a flat image; what the person was looking at can be extracted from this image.
In experiments, it was said that the system could successfully extract an image from the cornea so that one could tell what ball a person was aiming at on a pool table.
The results will be announced at the "SIGGRAPH" conference going on from August in Los Angeles.
Inspired by: http://www.yomiuri.co.jp/net/news/20040720ij11.htm [1]
Update: Thanks to sar7501 for the corrections to this article, shown in red.
|