|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Answer one: there are programs already that are specifically for human models
like Poser or MakeHuman and can be easily exported to pov using poseray
- http://poser.smithmicro.com/poser.html
- http://www.makehuman.org/
- https://sites.google.com/site/poseray/
Answer two: you must be kidding.
B. Gimeno
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 27.10.2012 03:41, schrieb Sven Littkowski:
> I propose to add a new shape (or object) to the existing POV-Ray shapes: Human.
> The "Human" would be controlled by parameters and allow to add more perspective,
> dimension or simply comparison to our scenes.
No problem - if you figure out the mathematical definition of that
shape, we'll figure out a way to implement it... :-P
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Sven Littkowski" <nomail@nomail> wrote:
> Hi,
>
> I propose to add a new shape (or object) to the existing POV-Ray shapes: Human.
> The "Human" would be controlled by parameters and allow to add more perspective,
> dimension or simply comparison to our scenes.
>
> A quality parameter (1 to 10) would control if the human is just a basic figure,
> like those wooden figures, or with all details.
>
> A next parameter defines the race which influences face and shape features.
>
> A next parameter defines fatness or meagerness.
>
> The next parameter defines the sex.
>
> There is one parameter for the height of the human. Needed, since positions
> (next parameter) could make it harder to set the size by scaling.
>
> A final parameter defines a few basic positions, such as standing, sitting,
> laying on belly or back.
>
> That human is just of one basic color for all the skin, another for the hair,
> and tree for shoes, trouser and shirt. Could be parameters, too.
Impossible. You name a couple of parameters, but there are hundreds of them. A
head can look up/down (elevation) and left/right (azimuth). An upper arm
(shoulder joint) also has 2 degrees of freedom. A hand (including wrist) has
23(!) degrees of freedom. Combine that with features like nose size and shape,
and you end up with trillions of combinations (even if the 7 billion people on
earth are different there's still a myriad of other people possible.)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
we need a "G.O.D." computer for this. I think you'll need a thousand or more
parameters to controll such a human "object"
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 23 Jan 2013 16:52:16 -0500, BertvdB wrote:
> we need a "G.O.D." computer for this. I think you'll need a thousand or
> more parameters to controll such a human "object"
Or http://www.makehuman.org - exports to OBJ format, which IIRC POV-Ray
can use.
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
How about just a *simple* human male figure (not posable, not super-realistic,
no triangle meshes, and only one color)? Not exactly what you have in mind(!),
but it would be a good stand-in or 'avatar' for giving scenes some interest and
scale reference.
I started working on such a thing several years ago--just a fast-rendering CSG
construction--but never got around to posting it. And at the time, it looked a
wee bit...laughable ;-) But I've updated it since then, so it looks a bit more
realistic.
The idea of making it posable had occurred to me--via a macro--but the sheer
*number* of controls needed would make the macro almost impossible to deal with.
So it's simply a static figure.
Just wondering if that would serve a useful purpose.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Thanks for all the answers.
By linking to other programs which can do this, it is indirectly admitted that
it could be possible. How else would otherwise those other programs manage
that..?
I remember, there was a time POV-Ray's scene code wasn't yet a fully functional
programming language. Just when the POV-Ray team added the MACRO feature to it,
it became more than it was up to then. That I have in mind now, too. There are
ways, so let's forget the limits. Best evidence are all those programs the
answerers linked to by themselves.
A basic figure would do, for now. It should look like a human, true, but it is
meant as a placeholder. Something to fill scenes with, to populate them, to make
them alive. But for anything really professional, still the software of other
companies might have to be used. I believe, it can be done. Something that looks
human from a certain distance, some archetype.
I just mean to motivate, to inspire. Not to ignore critics. :-)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 01.02.2013 05:34, schrieb Sven Littkowski:
> Thanks for all the answers.
>
> By linking to other programs which can do this, it is indirectly admitted that
> it could be possible. How else would otherwise those other programs manage
> that..?
>
> I remember, there was a time POV-Ray's scene code wasn't yet a fully functional
> programming language. Just when the POV-Ray team added the MACRO feature to it,
> it became more than it was up to then. That I have in mind now, too. There are
> ways, so let's forget the limits. Best evidence are all those programs the
> answerers linked to by themselves.
>
> A basic figure would do, for now. It should look like a human, true, but it is
> meant as a placeholder. Something to fill scenes with, to populate them, to make
> them alive. But for anything really professional, still the software of other
> companies might have to be used. I believe, it can be done. Something that looks
> human from a certain distance, some archetype.
>
> I just mean to motivate, to inspire. Not to ignore critics. :-)
We will /never/ have a hard-coded human figure shape in POV-Ray.
Here's why:
- It has been mentioned before that the geometry of even a simple human
stand-in would be pretty complicated, and can't be captured by just one
or two simple mathematical formulae. As a consequence, we would have to
resort to composing it from other primitives.
- The SDL has all the expressiveness it needs to achieve any possible
composition of primitives that the rendering engine supports, without
any rendering speed penalty. Where parameterization is desired, a #macro
can get you there.
- Hard-coding a human stand-in, rather than defining it via an SDL
macro, would rob the user of the possibility to modify the figure to
suit their special needs.
Thus, if ever a human stand-in should find its way into an official
POV-Ray release it will be in the form of an .inc file. (The same, by
the way, also goes for shapes of particular fruits.)
Actually that's how other software packages do it as well: Their
primitives are meshes, deformers, and some such. No human figures in
there. They are provided instead as external files, which tell them how
to compose a human model from those meshes, deformers etc.
This leaves us solely with the question why POV-Ray doesn't include a
human stand-in .inc, and whether it ever will.
To answer this question it is important to note that the human figures
provided by Poser, DAZ Studio, MakeHuman etc. are designed by 3D
artists, not software developers.
Now the POV-Ray dev team is a team of software developers. Our software
development skills far exceed our 3D modeling skills, so that's where we
invest virtually all of the time we can spare to get POV-Ray forward.
It's not that we wouldn't like to have a mannekin or something like that
among our standard include files. But I guess such a puppet would
ultimately have to come from the POV-Ray community, or we'll never have one.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
clipka <ano### [at] anonymousorg> wrote:
>
> It's not that we wouldn't like to have a mannekin or something like that
> among our standard include files. But I guess such a puppet would
> ultimately have to come from the POV-Ray community, or we'll never have one.
There are still the not very detailed blobman includes. A link to it is at
http://www.povray.org/resources/links/POV-Ray_Include_Macro_and_Object_Files/Macro_Files/
(Hope this works now, otherwise the link is somewhere at
POVray.org/Ressources...). Blobmans are simple but by studying the code one can
see
how many parameters are needed to pose only this simple figures.
But for small figures they may suffice.
Best regards,
ichael
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Thu, 31 Jan 2013 23:34:59 -0500, Sven Littkowski wrote:
> By linking to other programs which can do this, it is indirectly
> admitted that it could be possible. How else would otherwise those other
> programs manage that..?
A better - and maybe more relevant question - why reinvent the wheel?
Jim
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|