Returns the intrinsics matrix of the most recent semantic segmentation prediction. Contains values for the camera's focal length and principal point. Converts from world coordinates relative to the camera to image space, with the x- and y-coordinates expressed in pixels, scaled by the z-value.
for the camera's focal length and principal point. Converts from world coordinates relative to the
camera to image space, with the x- and y-coordinates expressed in pixels, scaled by the z-value.