0
$\begingroup$

I'm trying to make a graphical demonstration of how MeanShift works. I have 6 points in a collection, and a geometry node setup to apply a kernel based on the distance of these points from a plane position, then sum all these kernel results. Currently I just do this manually for each point. (see below) Is there some way I could use the Collection Info instances as the source of these positions instead of manually referencing every Object Info?

Nodes

Here's what the result looks like btw

enter image description here enter image description here

Improve this question
$\endgroup$
3
  • $\begingroup$ Collection info node, separate instances, attribute statistic node, position, min. I think. Except a statistic is for a single value, maybe you could convert instances to points and then do geometry proximity to points. $\endgroup$
    – Markus von Broady
    Commented Apr 27, 2023 at 23:27
  • $\begingroup$ I'd initially used the Geometry Proximity node, except that's not correct because it will only give the distance from the nearest point. I need to sum the weighted distance from every point. $\endgroup$
    – Adam Dernis
    Commented Apr 27, 2023 at 23:40
  • $\begingroup$ you could use attribute statistics node and use the sum, or use the accumulate field node and use the total. $\endgroup$
    – shmuel
    Commented Apr 28, 2023 at 21:09

1 Answer 1

Reset to default
2
$\begingroup$

I don't know what you're doing in your custom groups, which may be crucial for the problem, but if you wanted to sum all locations, you would do it like so:

Improve this answer
$\endgroup$
1
  • $\begingroup$ This looks like a good step, as I do need to sum them. But first I need to find the distance from the point and run it through a Kernel. This is certainly cool and useful though! $\endgroup$
    – Adam Dernis
    Commented Apr 30, 2023 at 0:13

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .