Here is the problem:
There is a number of panoramic images available, made mostly in the indoor environment. Our user is assembling them all into a virtual tour, where a different kind of user will be traveling from one image to another by clicking on some sort of hotspot in a panoramic image (e.g. like Street View in Google Maps, but much smaller scale).
So, one way or another, our user will need to create connections between panoramas, with each connection also creating two hotspots - one for each image. Because the environment is indoors, the only images/locations to be connected are those where it's possible to physically go from one location to the other (and back - assume all connections are two-sided).
Unfortunately, no coordinates where the images were taken are available for images. We can assume that most of the images/locations are taken on more or less the same height level, so only the horizontal positioning matters. Orientation data, however, may or may not be available - we are still looking into it.
The question is: how to implement the UI and workflow for our user so they can create those connections easily and efficiently?
At this point, only the basic suggestions are needed. Is it possible to avoid creating every connection manually? Is it necessary to make the user create a map of sorts? Has this been done before and are there good examples?