I am trying to extract the motion field from the visionworks example. i saw the post in the nvidia forum here but i still do not understand how to get the motion field.
The documentation says that the motion fiels is a vx_image with type NVX_DF_IMAGE2F32, this means there are two channels.
Someone know the meaning of each channel? I even realize that the size of the image is the half of the input images used to detect the movement.
In the nvidia post the moderator said " that value(x, y) is the motion of the point (x, y)", but i do not understand what this means.
I understand the concept explained "More precisely, for point(x, y) with field value(mx, my) in frame N. The matched point in the frame N-1 is (x+mx, y+my)."
but i could not match the result i get in the vx_image with these parameters.