Hello,
I’ve got multiple cameras which view the same object. They each estimate the rotation and location of the object with SolvePNP.
I know the location of the cameras and the object, and I know how to combine location data from multiple cameras to have the best result.
My issue I that I don’t know how I can do the same for rotation. As far as I know, the best rotation is the one on the plane perpendicular to the vector from the camera to the object.
Is there a way or a formula to combine the rotation along these best axes ?
1 Like