So I have a list of points (2d) that should be inside a polygon, and a second list of points that should be outside the polygon. I want to find the polygon that encloses the "inside" points while excluding the "outside" points.
The resulting polygon will almost certainly be concave. I would like to minimise the number of edges it has, but any other sensible criteria would probably do. It seems likely that some approach that started with the convex hull, and then took chunks out of it would work.
I am wondering if this could be achieved using alpha shapes, while varying alpha. There is already a nice python implementation for that; https://pypi.org/project/alphashape/ but I haven't worked out exactly how I'd need to vary alpha.