0

I'm new to vector math, I've mostly done plane / line intersections.

I'm currently trying to do a very simple stereo vision, and the idea is to use a pyramid (representing a feature being projected from a camera), and store it in an octal space tree (in this case represented by points).

So I'm trying to figure out how you could tell if a point is inside of a pyramid.

I thought of simply having 4 or 5 planes representing the pyramid, and then checking on which side the point is by getting it's signed distance.

But is there a better way?

LemonJumps
  • 1
  • 1
    You can consider barycentric coordinates, https://en.wikipedia.org/wiki/Barycentric_coordinate_system#Barycentric_coordinates_on_tetrahedra – Randy Marsh Jul 11 '24 at 18:11
  • Why octal space tree? Signed distance is ok, see this for example: https://math.stackexchange.com/questions/4216084/check-if-3d-point-is-inside-square-based-pyramid or this https://stackoverflow.com/questions/68641598/check-if-a-3d-point-is-in-a-square-based-pyramid-or-not – z100 Jul 11 '24 at 18:20
  • @z100 I want to use it mainly for it's scalability and speed, I'm expecting to deal with 1Million+ features in total. That said, I can still use SDFs with it too. – LemonJumps Jul 12 '24 at 05:45

0 Answers0