Suppose I have a prior knowledge of the movement of 4 points (in target coordinate system) from every frame n to frame n+1.
Given 4 correspondence points I can find the projective matrix from frame n to frame n+1 (my texture is 2D, no depth required).
In addition, using a trackable marker I can also obtain the pose matrix using trackable->getPose()
I would like to find a "new" pose matrix which is in a way better and more accurate than the original based on the projective matrix that I found. How exactly should I interoperate the getPose() coefficient in terms of coordinate system? And what would be a good approach to "set" the pose matrix entries?