Computer Vision System Toolbox has the functions addView and addConnection. Both functions can accept orientation and location information through name-value pair arguments. The online reference mentions that both arguments are "relative to the first camera". However, addView function refers to adding a single camera view while addConnection actually connects two camera views.
Since addView has only a single camera view, I would assume that its orientation and location is relative to the first view added to a view set, not to the latest (previous) view added. Is this a correct assumption?
On the other hand, addConnection has two view IDs as input arguments and it is confusing when the online reference states "relative to the first camera". Does this mean relative to the first ever view added to a view set? Or relative to the first view ID parameter of this function? To state the question more clearly, let us assume we have 3 views in a viewSet object and we are adding a connection between views 2 and 3. Are the orientation and location parameters of the addConnection function relative to view with viewID1 which, in this case, is view 2 or relative to the very first view added to the view set (view 1)?
It seems to me that the same text ("orientation of the second camera relative to the first") is used in the online reference to mean different things for the two functions. For addView it seems to mean relative to the very first view added to a view set and for addConnection it seems to mean relative to the first viewID argument supplied to the function? A clarification would be appreciated...