- This topic has 3 replies, 2 voices, and was last updated 9 years ago by
.
- Posts
In the development guide of Gaze SDK, it is mentioned that there is a concise way to visualise the calibration result using Tobii calibration plot visualisation.
It sounds like to me that it is some function that I can readily call from the library, probably from the Calibration class. Is this right or not?
I actually got a follow-up question with that:)
The reason that I would like to visualize the calibration results is that I want the participant / operator to be able to check the calibration result visually and manually. What I want to do is as follow:
- – Visualize calibration result (for checking quality manually)
- – Remove calibration points of bad quality (using calibration_remove_point)
- – Recalibrate those points (using calibration_add_point)
- – Recompute the calibration with updated calibration data (using calibration_compute_and_set)
- – loop back to the first step until good quality of calibration is achieved
These are the logic that I can imagine using Gaze SDK to realise what I want to do. Please correct me if anything is wrong.
Any kind of input would be appreciated. Thanks!!Hi Gigi,
The Tobii Gaze SDK does not include a tool or code sample for visualization of the calibration result, but it can provide you with the data you need to create one yourself.
The developer’s guide includes an image of how a calibration plot visualization can look, and the image is based on the kind of calibration plot visualization that is familiar for many users of certain Tobii software products.
To create a similar visualization you need to get the newly computed and set calibration from the eye tracker and then extract the necessary data for each calibration point.
In the Gaze SDK for .Net you can retrieve a calibration from the eye tracker as a Calibration object using either the GetCalibration or the GetCalibrationAsync method. On the Calibration object you can call GetCalibrationPointDataItems to get an IEnumerable of CalibrationaPointData structs that gives you the information you need for each calibration point to create a visualization and the kind of re-calibration functionality that you listed above.
Here is a list of the available data in the CalibrationPointData struct:
- TruePosition – The point on the screen where the calibration stimulus was displayed.
- LeftMapPosition – The left eye’s gaze point on the screen after calibration.
- RightMapPosition – The right eye’s gaze point on the screen after calibration.
- LeftStatus – Information about the validity and usage of the left eye data.
- RightStatus – Information about the validity and usage of the right eye data.
You can create a calibration plot visualization like the one in the developer’s guide by calculating and drawing the offset vectors between the true position and the mapped position of each eye.
You can use the status information to indicate in the GUI to the participant/operator which calibration point needs re-calibration.
Please, take a look also at the WpfCalibrationSample for an example how to create a GUI for a calibration procedure.
- You must be logged in to reply to this topic.