[Bf-committers] Movie distortion algorithm

[Jonathan Merritt]

Well, the transformation between image and (xd, yd) is linear, so that
part's easy.  The other transformation between (xd, yd) and (x, y) is
non-linear, and looks quite painful.

Since the solution is not radially symmetric, there isn't a solution in
terms of r.  Hence, I would be tempted to write the terms for (x,y) => (xd,
yd) fully in terms of x and y, expanding the powers of r, and then attempt
to brute-force a solution.  You could try this with open-source
computer-algebra software such as Maxima.

Another alternative would be to use Newton's method (iterative) to find (x,
y) given (xd, yd).  This should be fairly easy because you can write the
Jacobian of the system symbolically.

If I have time, I'll have a go at one or both of the above.  What is your
time-frame for this?  It might take me a week to get around to it, so
anyone else should jump in.

Jonathan Merritt.

On Mon, Apr 16, 2012 at 4:06 PM, Dan Eicher <dan at trollwerks.org> wrote:

> These guys describe a fast preview mode (if that's the problem):
>
> http://www.vassg.hu/pdf/vass_gg_2003_lo.pdf
>
> ...not that I understand the math behind it.
>
> Dan
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers
>