Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

High Offset in advanced Calibration, duration of fixation point presentation #113

Open
paul-gallenkemper opened this issue Jan 17, 2025 · 1 comment

Comments

@paul-gallenkemper
Copy link

When runing our calibration + validation procedure, which is a five point calibration / four point validation, two-screen advanced calibration procedure using videos as fixation points designed to test infants, we regularly encounter offset values of over 5°. This also happens when we run the calibration on adults, but not when we run the inbuild calibration in TobiiProLab. Note that we have carefully set up the eyetracker in TobiiProManager on both PCs.
We first thought of increasing the time the fixation points are presented and used these parameters:

settings.UI.advcal.gazeHistoryDuration = 3;
settings.advcal.cal.paceDuration = 3;
settings.advcal.val.paceDuration = 3;
settings.advcal.collectDuration = 3;

As i couldn't find any detailed documentation on these settings yet, i' not sure which of these is actually useful. Even though the points are presented for longer, the offset remains high. When observing the gaze points on the administration display during the procedure, it becomes clear that the gaze point estimations are about 10-15 centimeters apart from each other per eye and this distance does not reduce with time.

Another issue with this longer presentation of the fixation points is also that the videos don't load immediately, resulting in stationary images being shown for a while before the motion sets in and other kinds of delays with the videos sound.

What could be possible causes of this gaze point seperation per eye? What are other parameters of the advanced calibration procedure we could adjust to solve this?

@dcnieho
Copy link
Owner

dcnieho commented Jan 17, 2025

Hi Paul,

This issue surprises me and is not something i have ever encountered. Could you:

  1. take a video recording of the operator screen during a whole such a run, so i can see exactly what happens?
  2. send the exact script you used for that video (you can cut out everything after calibration, but please do verify that it still produces the problem)
  3. also include the exact videos that you have problems with.

and email that all to me?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants