I am using Pupil Labs binocular eye tracker for recording different eye movements. According to the documentation, the gaze position and fixation position are given in normalized x, y screen coordinates (if the calibrated area is the screen).
I have two questions:
1. Will calculating eye movements measures that include fixation position such as Area of Interest (AOI) based measurments still be accurate?
2. How can I determine AOI in this case, since AOI will reflect a part of visual stimuli dispalyed on monitor (in pixel) ? I read through their open source code, and still didnt get how they normalized the coordinates.
Is there a standard method for normalization to follow?
Jihad, I do not know this device, but what I read in the Pupil Labs Wiki seems fairly clear:
"We use a normalized coordinate system with the origin 0,0 at the bottom left and 1,1 at the top right" (https://github.com/pupil-labs/pupil/wiki/Data-Format)
Even if the the data format within the eye tracking software was only single precision the numerical accuracy would be 0.000001, i.e. one could distinguish one million view points in each direction. Note that a full hd display has roughly two thousand (sorry I had written "million" before my edit) pixels in each row so that even in the worst case of only single precision one would be able to decribe the view point on the display within two thousands of a pixel.
Given the the measurement accuracy of only 0.6 deg of the device (the display will subtend no more than 30 deg with a properly positioned subject) I think there is no need to worry anyway.
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Agronomy
- Crop Science
- Crop