I have been working on a Unity project which uses a vuforia camera to lock a city layout to an image. However, my main issue is a typical one with image targets in that when there isn't enough detail from the original target in frame the model disappears.
My idea is to use a vuforia camera to identify a specific target, but then lock the model into place using a parallel ARCore camera in order that I can use the orientation sensors of the device to lock the city in position.
I am sure this is possible, as there is an example on the experiments with google that achieves this, but I can't seem to find any technical details on the practicalities. Does anyone know how this can be achieved?
Example here: