[Bf-committers] GSoC 2011: Improving Retopology Tools

[Sergey Kurdakov]

Hi Tom,

>A common method for retopology is over sculpted meshes.  You might
>want to retopologize creating only a small number of polys over an
>area with thousands, tens of thousands or more polys  (Right now you
>can do over 40 million polys in a single mutires mesh).  (with a poly
>budget of 150 for a simple object -  40 million/150 is about 250,000
>polys per quad of your retopo mesh)

somehow there should be decision - either your script could provide
that functionality or not
cleary it could not - it will create relatively small patches.

now Matt told that re topology will be to simplify some areas - not
all the case you mention.

in this case rohith approach will suffice and provide good result.

now big meshes - Nick applied a patch which effectively and quickly
makes remeshing over huge meshes.


> Wasn't claiming to :)

you was claiming that what I say is completely unappropriate and this
is cleary not a case.

we revealed a problem with rohith approach - only limited areas, but
due to Matt note and common sence it will suffice.
it could start from retopo of areas, then if good sparse matrix lib is
found ( and there are such libs ) it could be extended.

so the question is here - to select what is best on number of criteria.

obviously you script won't handle large areas to retopo and on small
meshes rohith approach has a hand - it preserves topology of mesh.

> While it is true that there might be some use in extending the retopo
> tools beyond the scope that was suggested for GSoC, the original
> inquiry was more narrow in nature.

the applicant must win an application so it should be interesting and
not trivial.

Regards
Sergey