dcnieho
commented
4 years ago For almost all Tobii eye trackers (except perhaps the new Nano and Fusion), the positioning stream should give you the same information as the 3D positioning info in the gaze samples. Its kinda like a scaled trackbox position if i understand correctly. And indeed, that stream does not get time stamped. Does it allow you to distinguish blink vs loss of face? One way to know that (at 5 Hz) is to look at the eye image stream (for the Spectrum at least), you get different resolution images when tracking face vs when not
On Tue, Aug 11, 2020 at 11:26 AM Ian Max Andolina notifications@github.com wrote:
We have problems where sometimes the face position is lost, with our subject still trying to do a task. This is an error of the Tobii, rather than the subject. At the moment we are only using the gaze stream, but valid/invalid cannot distinguish between an .e.g. blink vs the loss of the face resulting in loss of eye signal. Is the resolution of this stream the same as the gaze data, it doesn't get timestamped correct?
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/dcnieho/Titta/issues/26, or unsubscribe https://github.com/notifications/unsubscribe-auth/AANUOGKZYF3DKZBUHWFRTK3SAEFDTANCNFSM4P22TMVA .
iandol
We have problems where sometimes the face position is lost, with our subject still trying to do a task. This is an error of the Tobii, rather than the subject. At the moment we are only using the gaze stream, but valid/invalid cannot distinguish between an .e.g. blink vs the loss of the face resulting in loss of eye signal. Is the resolution of this stream the same as the gaze data, it doesn't get timestamped correct?