Re: [Algorithms] Combining normal maps

[<cas...@ya...>]

Tom Forsyth wrote:
> Gagh - I don't think there's any simple way to get back from a normal map
to
> a heightfield. 2D integration - yuk.

Another posibility is to extract displacement maps from the object, instead
of normals. If you just compute the distances from the low poly to the high
poly it will look bad, because the low poly surface isn't continuous. Maybe
it would be possible to interpolate position and normals in a different way,
using n-patch interpolation or something similar. Then you can process the
resulting heighfields with ease. And you can directly apply them to an
n-patch model on the parphelia! However, there are some issues with hard
edges that I don't how to deal with.

btw, I don't also know how does the n-patch capable hardware interpolate
other parameters appart from the normals. Should binormals and tangents be
interpolated quadratically or a linear interpolation is enough? Should the
tangent basis be renormalized?


Ignacio Castaño
cas...@ya...

_______________________________________________________________
Copa del Mundo de la FIFA 2002
El único lugar de Internet con vídeos de los 64 partidos. 
¡Apúntante ya! en http://fifaworldcup.yahoo.com/fc/es/