h a l f b a k e r yA riddle wrapped in a mystery inside a rich, flaky crust
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
This is a system where a computer tries to guess where a person is looking by detecting their pupil. Now, the human pupil is black in visible light, but reflective in infrared. This system has two cameras, one visible and one infrared. It shines an infared source in front of the cameras. The imaging
software tries to find patches of color that are black in one image and infrared in the other.
One of the better systems on the market
http://www.eyelinkinfo.com/index.php
+also tracks reflections of light on the cornea [loonquawl, Feb 26 2009]
Wikipedia: Eye tracking: Technologies and techniques
http://en.wikipedia...gies_and_techniques
"Most modern eye-trackers use contrast to locate the center of the pupil and use infrared and near-infrared non-collimated light to create a corneal reflection (CR). The vector between these two features can be used to compute gaze intersection with a surface after a simple calibration for an individual." [jutta, Feb 26 2009]
[link]
|
| |
That is exactly how video-based eyetrackers work. |
|
| |
What about glasses or contact lenses? |
|
| |
contact lenses tend to spoof the (normally more exact) corneal reflection, because that relies on the slightly bent (relative to the eye) surface of the cornea, and the contact lenses, while having this curvature too, tend to slip around a little during fast eye movements. |
|
| |
The algorithm searching for the pupil (reflective round spot somewhere in the picture) is pretty robust, though. The infrared is ~800nm, so the glasses are not opaque to the wavelength (as some would be from 1000nm onwards) |
|
| |