|
 |
"Bald Eagle" <cre### [at] netscape net> wrote:
> That approach was also close to what I was doing, however it doesn't match the
> curvature of the spline segments at the endpoints.
>
> Visually, it still looked too flat-then rounded-then flat-then rounded....
>
> Although I didn't pore over your code yet, I quickly skimmed it to see where the
> control points were calculated, and then bolted on my curvature comb code to
> visualize the curvature over all of the spline segments.
>
> You can see where the mismatches are.
>
> So I still think I have to degree elevate and then I can mirror the additional
> control points to get the curvature to match.
>
> (Also, FYI: the point/tip is actually a hard point. So 2 separate composite
> splines - 1 on the left and 1 on the right)
>
> I'll see if I have enough time to patch in my degree elevation code and extend
> the control-point mirroring that one extra "level" into the splines.
It is interesting to see the discontinuities visualized like this.
Have you also made similar tools for working with curved surfaces?
--
Tor Olav
http://subcube.com
https://github.com/t-o-k
Post a reply to this message
|
|
|
|
"Tor Olav Kristensen" <tor### [at] TOBEREMOVEDgmailcom> wrote:
> It is interesting to see the discontinuities visualized like this.
>
> Have you also made similar tools for working with curved surfaces?
That was what the whole Gaussian curvature project was about.
Iterating over a surface and showing the curvature combs would be - messy.
I also covered the various production methods of visualizing curvature and
discontinuities using zebra stripes, dots, isophotes, etc.
There is, of course, always so much more to do with this stuff.
- BW
Post a reply to this message
|
|
