Hi Neil, thanks for the help. Yes, I used ray casting and it seemed like the positions projected in the Unity project correspond to the actual gaze directions. However, I cannot discover where in the data I can find these positions.

Based on the *developer docs * it seems like I should use GazeDirection and GazeDistance rather than gaze_point_3d. However, the exported files from Pupil Capture don't show these data. Is there a way I can export GazeDirection and GazeDistance data?

Hi @user-8bce45 ! Pupil Capture does not have knowledge of the VR environment, even though you can cast the screen, it will not record the 3D environment.

So essentially, if you are doing ray-casting as you mentioned above. You can basically get the "hit point" coordinate in Unity.

Then you will need to store it within Unity to a csv file, together with the timestamp.

Since you mentioned that you want to know what has been gazed, you can use the raycast and colliders in your objects of interest, such that you can detect whenever your gaze vector "hits" an object. https://docs.unity3d.com/2018.3/Documentation/ScriptReference/RaycastHit.html

Hi! I'm currently using the Screencast Prefab attached to the Main Camera to my XR Origin to share the view into Pupil Capture. However, I found the the script causes double vision within the headset.

I am unsure how to solve this as I've followed what was mentioned in the hmd-eyes document: Make sure the camera uses the VR head transform as the origin - normally just by adding it as a child to the main/VR camera - and has position and rotation set to zero

edit: I've just read the following point about performance, this might be a related matter.