|
 |
"Edee" <rah### [at] yahoo com> wrote:
> I am putting together an architectural concept structure that requires
> adding some human images to provide some perspective on the size of the
> structure. Are there any sample pov files out there that I can use?
> thanks in advance.
Hi,
I've done this using three different approaches.
First, paste a human image onto a thin transparent box that faces the
camera. This approach is pretty straightforward, assuming you have a good
image to start with. I found some freebies at http://www.got3d.com/ If you
look around their store, each of their packages usually has a free sample.
The problem with this technique is that the lighting/shadow on the image
typically won't match the lighting of the scene. In addition, you need lots
of images to avoid repetition if you have several humans in your scene. I
think this approach would work for you, but you may wish to consider a
couple of others:
Second, you could make a CSG human - you don't need much detail if your
humans are far from the camera. You can randomly scale the humans, randomly
colour their hair, skin, clothes, etc. Easy to get lots of variety - and
the lighting/shadow works correctly.
Third, you could use a 3d human model (best if you need realism/closeups of
the humans or if you need them positioned just so). I highly recommend the
free human models from DAZ
(http://www.daz3d.com/shop.php?op=catmain&cat=6). These models can be posed
(joints positioned) in tools like Poser or DAZ|Studio (free!), exported to
OBJ, and converted to POV-readable formats like mesh or mesh2 by POSERAY
(free!). A lot of work sometimes, but can be very satisfying. This method
generally takes more memory than the other two, but can be efficient if you
have many instances of the same model.
I have a few examples of images using methods #1 and #3 at
http://fitzel.ca/dart/povray/metro/metro.html
Good Luck,
Rob.
Post a reply to this message
|
|
|
|
"RobF" <nomail@nomail> schreef in bericht
news:web.43a4efb8b8523b6629c50a910@news.povray.org...
>
> First, paste a human image onto a thin transparent box that faces the
> camera....
Instead of a box, it is often better even to use a polygon as it has no
third dimension. Use an image_map for the human figure with an alpha channel
for the transparency around the outline. Be aware also that if you use a
fog, the transparant part may become visible!
Thomas
Post a reply to this message
|
|
|
|
"Thomas de Groot" <t.d### [at] internlnet> wrote:
> "RobF" <nomail@nomail> schreef in bericht
> news:web.43a4efb8b8523b6629c50a910@news.povray.org...
> >
> > First, paste a human image onto a thin transparent box that faces the
> > camera....
>
> Instead of a box, it is often better even to use a polygon as it has no
> third dimension. Use an image_map for the human figure with an alpha channel
> for the transparency around the outline. Be aware also that if you use a
> fog, the transparant part may become visible!
>
> Thomas T
Thanks all for the input, I have downloaded DAZ and but plan short term to
just use the polygon idea and then move on to using DAZ as time allows for
future projects.
Post a reply to this message
|
|
