[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gzz] 9th (jvk)
From: |
Tuomas Lukka |
Subject: |
Re: [Gzz] 9th (jvk) |
Date: |
Mon, 9 Sep 2002 18:32:40 +0300 |
User-agent: |
Mutt/1.4i |
On Mon, Sep 09, 2002 at 06:14:37PM +0300, Janne Kujala wrote:
> 9th: Mostly libpaper work
>
> Added "vertex" entry point to PaperPass, allowing
> nonlinear custom mapping between object and paper coordinates.
>
> Only simple TexGens are currently supported.
>
> For EmbossTexGen, there are some questions:
> - should we distort the paper or the lighted image of the paper?
Either: the point was that you can pass lighting coordinates for
each point as well, if desired. If none are passed, then just
distort lighted image.
> - can I assume that the paper is still flat?
No, but that is taken care for lighting by the above answer.
> - how will the application compute the object and paper
> coordinates -- what kind of derivates (for the local
> linearization) can be expected to be easily computable?
Ouch. That's a nasty one. For the simple case, assuming distortion of
lighted image, this is not a problem. However, for distortion, the
derivatives do need to be passed.
You can make the more complicated call a placeholder for now, we won't
need that for the demos.
Tuomas
- [Gzz] 9th (jvk), Janne Kujala, 2002/09/09
- Re: [Gzz] 9th (jvk),
Tuomas Lukka <=