"Warp" wrote:
>   You could calculate the average of the normalized
> cross-products of all pairs of vectors.
>
>   If there's any word in that sentence which you didn't
> understand, drop me a line and I'll explain (I can't
> know how well you know the terminology).

Thanks, I understand it perfectly, and I've been thinking of doing that
myself, but it sounded inefficient and slow to me.

However, I realize now that for n points there's just n*(n-1)/2 pairs I
think, which is not too bad after all for the amounts of numbers I have.

So, how do this problem with the vectors relate to the problem with finding
the normal of a plane that best fits a bunch of points? Well, I figure that
when I have that bunch of points I can find the average point and look at
the vectors that go from that point to all the other points. Using the
method with the vectors I can find the normal of the plane, and I also know
that the plane goes through the average point.

For the curious, I need this for my spider walking animation, where I need
to find a good "up vector" for the spider dependent on where its feet points
are.

Rune
--
3D images and anims, include files, tutorials and more:
Rune's World:    http://rsj.mobilixnet.dk (updated June 26)
POV-Ray Users:   http://rsj.mobilixnet.dk/povrayusers/
POV-Ray Webring: http://webring.povray.co.uk

Post a reply to this message