[Solved] Changing the duration or the amount of data collected for each calibration point

Home Forums Legacy SDKs [Solved] Changing the duration or the amount of data collected for each calibration point

Tagged: 

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • #4012
    Ozan Saltuk
    Participant

    Hi.

    I need to know if it is possible to change the duration or the amount of data the eye tracker collects for each calibration point. I want to experiment in calibration via pursuits, and I had the idea to make the time the system takes for calibration points faster. Is there any way to do this? I think I need to modify the tobiigaze_calibration_add_point_async() function. Do I need to somehow decompile and change the .dll file? Is such thing allowed? Is there a better way of achieving this? (I am using the C/C++ libraries)

    Thanks in advance.

    #4013
    Anonymous
    Inactive

    Hi,
    I have been looking at this idea in the past, and I remember that the shortest time was not short enough to allow for calibration using moving objects. I do not know how many “samples” the calibration algorithm needs to fit the mathematical models and if it breaks if the samples are too far apart or too few. Hopefully someone from Tobii can give you a good answer/solution 🙂

    best

    #4022
    Ozan Saltuk
    Participant

    Thanks for the answer. That’s too bad, but I hope there is a way to do it(and someone from tobii can shed some more light 🙂 ). My second idea would have been to manipulate the calibration after the calibration is completed but I read that that is impossible.

    #4031
    Jenny [Tobii]
    Participant

    Hi Ozan.

    Francisco is right. The eye tracker needs enough time to collect data for a specific stationary point before moving to the next.

    What you want to do would require manipulation at the core of our proprietary algorithms for calibration and mathematical model of the eyes. It is not possible. And it is not possible to manipulate the calibration afterwards, due to the calibration format being proprietary as well, as it is tightly coupled to important aspects of the eye model.

Viewing 4 posts - 1 through 4 (of 4 total)
  • You must be logged in to reply to this topic.