If the nodes (Points) are “suitable” (a big thing that one) there’s various ways like using parallel *Ball Pivot (blame IBM Labs for that) or Marching Cubes, Crust, CoCone algos and the likes.
If on the other hand your nodes are not that “ideal” there’s also various ways. Yoo can also “wrap” your objects into a tri/quad Mesh using a PE thingy like K2, Here’s some test blobs wrapped by a quad Mesh (the blue) using K2 Methods
None of the above are achievable (even the K2 way) without code (mid to advanced Level). Not sure if there’s “add-ons” around that attempt to do that kind of thing (but someone did a BP C# in the past if memory serves well).
BTW: Search Internet on that topic (tons of stuff/Methods around - most in C++)
BTW: Here’s a classic BP (on “suitable” Points) C# solution In action:
Thank you for your kind response, I tried the ball pivot method and it did not yield good results. maybe K2 could offer some better reuslts. I guess I have to look into methods online and try to integrate that into GH.
As I said the “spread” of the points (i.e. are they “evenly” distributed ?) affects BP/MC results. Plus there’s the quality of the code (Bernardini et all just outlined the idea - implementaion is another animal). That said I’ve seen some free indicative attempts/solutions (in C++) that are full of various issues.
It’s most unlikely that you can find a free high-perf issue-free fast code/solution since the topic is rather hot (reverse engineering, self driving vehicles, defense drones mixed with AI etc etc) and thus … blah, blah. That said the best work these days on that matter is done in Israel (Intel got an almost start-up Tel Aviv based company for 17 billion).
Anyway if advanced coding is not your game … K2 (with stuff available as “components”) is your best “general case” option.