Is there a way to render stereo versions of an image in one pass? Other
than setting up some mirrors?

    I want to do an animation in stereo and I don't want to paste together
all those images...

Post a reply to this message

Bill,
I don't think there is a way. I render them seperately then use
a program called DepthCharge Developer Studio to batch process
the files. DepthCharge Developer Studio will output the stereo
views in anaglyph, interlaced, or side-by-side formats. I
process the separate files into interlaced stereo images then
convert to an AVI animation file, finally output the files as
digital video into my Digital-8 camcorder. It would be great
if some tallented programmer wrote a front-end to POV-ray to
do single pass stereo rendering! I hope it is a feature in
POV-Ray 4.0. I also use Hash Animation:Master to render animations
directly to stereoscopic video files.

Harolddd


"Bill DeWitt" <bde### [at] cflrrcom> wrote in message
news:3b28e996$1@news.povray.org...
>
>     Is there a way to render stereo versions of an image in one pass?
Other
> than setting up some mirrors?
>
>     I want to do an animation in stereo and I don't want to paste together
> all those images...
>
>

Post a reply to this message

"Harold Baize" <bai### [at] itsaucsfedu> wrote :
>
> Bill,
> I don't think there is a way.

    I was pretty sure but thought I should ask before building a stereo
camera macro... all that vector math that I don't really understand...
<shudder>

    Actually it should be pretty easy, just three boxes that are in the
right place and stay there when the camera move...

Post a reply to this message

Bill,
Can you help me understand what your macro would do?
The camera only needs to shift a little on the x
dimension between renders. Dosen't really need a lot
of math unless you're moving the camera in toward
objects, in which case you would need to change the
spacing between cameras to bring them closer together
as you get close to objects.

Clue me in. Thanks,
Harolddd

"Bill DeWitt" <bde### [at] cflrrcom> wrote in message
news:3b28ef7b$1@news.povray.org...
>
> "Harold Baize" <bai### [at] itsaucsfedu> wrote :
> >
> > Bill,
> > I don't think there is a way.
>
>     I was pretty sure but thought I should ask before building a stereo
> camera macro... all that vector math that I don't really understand...
> <shudder>
>
>     Actually it should be pretty easy, just three boxes that are in the
> right place and stay there when the camera move...
>
>

Post a reply to this message

"Harold Baize" <bai### [at] itsaucsfedu> wrote in message
news:3b28f2fe$1@news.povray.org...
> Bill,
> Can you help me understand what your macro would do?

    The idea is to get both views in one render. To do this, you have
mirrors that cut the scene in half. I would use boxes because they render
quickly. See PBI for cursory diagram.

Post a reply to this message

Yep, should work. In the real world we have
mirrors for 3D photography. They're often
called "beam-splitter" attachments for regular
35mm cameras. The term is a misnomer because the
beam of light is not split, the frame of film is
split. So you'd render at double width? Something
like 1600x600?

Harold

"Bill DeWitt" <bde### [at] cflrrcom> wrote in message
news:3b2908f6@news.povray.org...
>
> "Harold Baize" <bai### [at] itsaucsfedu> wrote in message
> news:3b28f2fe$1@news.povray.org...
> > Bill,
> > Can you help me understand what your macro would do?
>
>     The idea is to get both views in one render. To do this, you have
> mirrors that cut the scene in half. I would use boxes because they render
> quickly. See PBI for cursory diagram.
>
>

Post a reply to this message

"Harold Baize" <bai### [at] itsaucsfedu> wrote in message
news:3b290e40$1@news.povray.org...
>
> Yep, should work. In the real world we have
> mirrors for 3D photography. They're often
> called "beam-splitter" attachments for regular
> 35mm cameras. The term is a misnomer because the
> beam of light is not split, the frame of film is
> split.

    Exactly.

> So you'd render at double width? Something
> like 1600x600?

    Something like. The problem seems to be getting the secondary mirrors to
angle correctly to "look_at" the object. Vector math... I hate it... I'm
getting nowhere.

Post a reply to this message

Seems to me it would be the same as the physical
camera attachment, just four mirrors at 45 degree
angles.

 |    |
 |    |
 \-\/-/
   ||
   __

You don't need to converge on the object, if the
outer mirror separation distance is right. Almost
all stereo cameras, and most stereo photography
using two synchronized cameras, use parallel
lenses to avoid keystone distortion. Then again
I could be wrong. Keep me informed.

Harold


"Bill DeWitt" <bde### [at] cflrrcom> wrote in message
news:3b29105e$1@news.povray.org...
>
> "Harold Baize" <bai### [at] itsaucsfedu> wrote in message
> news:3b290e40$1@news.povray.org...
> >
> > Yep, should work. In the real world we have
> > mirrors for 3D photography. They're often
> > called "beam-splitter" attachments for regular
> > 35mm cameras. The term is a misnomer because the
> > beam of light is not split, the frame of film is
> > split.
>
>     Exactly.
>
> > So you'd render at double width? Something
> > like 1600x600?
>
>     Something like. The problem seems to be getting the secondary mirrors
to
> angle correctly to "look_at" the object. Vector math... I hate it... I'm
> getting nowhere.
>
>

Post a reply to this message

Bill,
I just looked at your diagram and I see what you
mean. Most people who work in stereo would draw
it differently, the lines coming from the object
would be parallel. Our eyes are basically parallel
when viewing things beyond about 6 feet. What you
present in your diagram is similar to what happens
when we look at something very close, like reading
a book. Our eyes converge.
If your objective is to represent a small object
close up, convergence will still not serve you well.
The reason is that most viewing systems present the
images to our eyes in a parallel fashion. As a result
images rendered from convergent perspectives seem
distorted and difficult to view.

Harold

"Bill DeWitt" <bde### [at] cflrrcom> wrote in message
news:3b29105e$1@news.povray.org...
>
> "Harold Baize" <bai### [at] itsaucsfedu> wrote in message
> news:3b290e40$1@news.povray.org...
> >
> > Yep, should work. In the real world we have
> > mirrors for 3D photography. They're often
> > called "beam-splitter" attachments for regular
> > 35mm cameras. The term is a misnomer because the
> > beam of light is not split, the frame of film is
> > split.
>
>     Exactly.
>
> > So you'd render at double width? Something
> > like 1600x600?
>
>     Something like. The problem seems to be getting the secondary mirrors
to
> angle correctly to "look_at" the object. Vector math... I hate it... I'm
> getting nowhere.
>
>

Post a reply to this message

"Harold Baize" <bai### [at] itsaucsfedu> wrote ;
>
> As a result
> images rendered from convergent perspectives seem
> distorted and difficult to view.

    That's what I have been finding out. I think I need to start over...

Post a reply to this message

"Harold Baize" <bai### [at] itsaucsfedu> wrote in message
news:3b291b2f$1@news.povray.org...
> Seems to me it would be the same as the physical
> camera attachment, just four mirrors at 45 degree
> angles.
>
>  |    |
>  |    |
>  \-\/-/
>    ||
>    __


    When I do that, I get a single image in the center, which
experimentation reveals is both halves joined in the middle.

    Life is hard when you don't know what you are doing.

Post a reply to this message

> > Seems to me it would be the same as the physical
> > camera attachment, just four mirrors at 45 degree
> > angles.
> >
> >  |    |
> >  |    |
> >  \-\/-/
> >    ||
> >    __
> 
> 
>     When I do that, I get a single image in the center, which
> experimentation reveals is both halves joined in the middle.
> 
>     Life is hard when you don't know what you are doing.

I messed around for a while (about a week or so on and off) trying to get 
this system to work, but got nowhere.  As far as I can tell, the problem 
is that the diagram above assumes an orthographic view.  Using an ortho 
camera, it might be possible to get it to work by tilting the two outer 
cubes so that their reflections are focused on a point in space...

But that's just speculation, and I haven't the faintest idea of how to do 
it with a perspective camera...

Bye for now,
     Jamie.

Post a reply to this message

Hmmm, so a virtual camera is different from a real
one? or is there something were missing here?
Here's an interesting idea, put transparent color
filters in front of the front mirrors and see if
the resulting image is an anaglyph! Another possibility
is that it needs a septum to keep the images seperate.

HB



> > Seems to me it would be the same as the physical
> > camera attachment, just four mirrors at 45 degree
> > angles.
> >
> >  |    |
> >  |    |
> >  \-\/-/
> >    ||
> >    __
>
>
>     When I do that, I get a single image in the center, which
> experimentation reveals is both halves joined in the middle.
>
>     Life is hard when you don't know what you are doing.
>
>
>

Post a reply to this message

"Harold Baize" <bai### [at] itsaucsfedu> wrote :
>
> Hmmm, so a virtual camera is different from a real
> one? or is there something were missing here?

    I would bet on the second. I tried orthographic to see if that fixed it
(thinking that it needed to have a little built in distance rather than a
'point' camera)... no go...

    I have a little bit of time to mess with it tonight, but I doubt that I
will figure it out...

Post a reply to this message

Why don't any one know this link?

http://sabix.etdv.ruhr-uni-bochum.de/sabpov/

If you want to render Stereo images, let me know it.

Let me also know, if ther is anyone of the theoretical stereo
camera macros is working well.

Stefan

Post a reply to this message

On Thu, 14 Jun 2001 12:42:59 -0400 Bill DeWitt wrote:

>
>    Is there a way to render stereo versions of an image in one pass? Other
>than setting up some mirrors?
>
>    I want to do an animation in stereo and I don't want to paste together
>all those images...
>

  Hi, Bill. See Glenn McCarter's "Dawn Patrol" image and text description. I
think this will do what you want.

http://www.irtc.org/stills/1998-02-28.html

-- 
Alan - ako### [at] povrayorg - a k o n g <at> p o v r a y <dot> o r g
http://www.povray.org - Home of the Persistence of Vision Ray Tracer

Post a reply to this message

Well damn, why didn't we hear about this!!
Great. Thanks for letting us know. I'd very
much like to have a copy of the program.
Any chance it could be added to MegaPOV or
POV 3.5??

Harold

"Benk" <Ste### [at] ruhr-uni-bochumde> wrote in message
news:3b299dc5$1@news.povray.org...
> Why don't any one know this link?
>
> http://sabix.etdv.ruhr-uni-bochum.de/sabpov/
>
> If you want to render Stereo images, let me know it.
>
> Let me also know, if ther is anyone of the theoretical stereo
> camera macros is working well.
>
> Stefan
>
>
>

Post a reply to this message

>>
>>  |    |
>>  |    |
>>  \-\/-/
>>    ||
>>    __
>
>
>    When I do that, I get a single image in the center, which
>experimentation reveals is both halves joined in the middle.

Ok, time for some physics...; let's just consider the right-hand image.

The ray that corresponds to the left side of the camera leaves in the +y
direction.  It reflects off the first mirror, and is now travelling <1,0>
It reflects from the second mirror, and is now travelling <0,1>.

I.e The left-hand edge of the right-hand image is the view dead ahead.

For a stereo image, if your angle of view is theta degrees, then you want
the left side of the right hand image to correspond to <0,1> rotated by
<- theta/2 degrees>.

Now consider the right side of the image.  With your mirror arrangement,
the ray leaves the camera at <sin(theta/2), cos(theta/2)>.  After the double
reflection, it'll leave the second mirror at that same angle.  So this ends
up at the right angle, although there is a bit of distortion introduced by
the fact that the right edge of the mirror is further away than the left edge.

What you want to have happen is that the mirror combo should reflect the
left-hand edge of the right-hand image to an angle of <-sin(theta/2), 
cos(theta/2)>; the right hand edge of the image to an angle of <+sin(th/2),
cos(th/2)>;  This is equivalent to the center of the right hand image 
looking dead ahead.

I.e. camera angle <+sin (theta/4), cos(theta/4)> gets reflected to <0,1>

Actually doing the math is fairly simple, but I don't happen to know what
the field of view for a standard camera is, so I can't run the numbers 
for you.

As a side note, If you use an arrangement more like:

    |
    |
/   |
||  |
| | |
|  ||
e   /

You'll have less distortion, because the difference in beam-length to the
different sides of the mirror will be less (although the total beam-length
will be longer)

Jamie

Post a reply to this message

"James Tonkin" <jrt### [at] calumcsclubuwaterlooca> wrote :
>
> Actually doing the math is fairly simple, but I don't happen to know what
> the field of view for a standard camera is, so I can't run the numbers
> for you.

    Actually just setting up the mirrors has been fairly simple. Making them
move with the camera as it pans and sweeps while keeping the mirrors pointed
at the look_at point is the hard part...

    There seems to be another problem with the point or perspective camera.
My drawing (in PBI) shows what I think is happening. I think I will end up
wanting a curved mirror or a different camera.

Post a reply to this message

in news:3b2a50bf$1@news.povray.org Bill DeWitt wrote:

>     Actually just setting up the mirrors has been fairly simple.

Havn't followed this thread, but maybe this helps:
http://www.irtc.org/stills/1998-02-28.html
The third-place picture is done using mirrors

Ingo

-- 
Photography: http://members.home.nl/ingoogni/
Pov-Ray    : http://members.home.nl/seed7/

Post a reply to this message

Bill DeWitt wrote:
> 
> Is there a way to render stereo versions of an image in one pass?
> Other than setting up some mirrors?
> 
> I want to do an animation in stereo and I don't want to paste together
> all those images...

Use POV-Ray to paste them together.  Paint one half of a plane with
one image, the other half with the other.  Use the str(,,) function to
select the file to use for each side, and then render as an animation.
Don't anti-alias the paste-up render.

I used a similar process to do the blocky digital censoring in my IRTC
entry titled "Showbots."

Regards,
John
-- 
ICQ: 46085459

Post a reply to this message

"ingo" <ing### [at] homenl> wrote in message
news:Xns### [at] povrayorg...
> in news:3b2a50bf$1@news.povray.org Bill DeWitt wrote:
>
> >     Actually just setting up the mirrors has been fairly simple.
>
> Havn't followed this thread, but maybe this helps:
> http://www.irtc.org/stills/1998-02-28.html
> The third-place picture is done using mirrors

    Thanks, but I am working on something a little beyond just a static
image. I am going through the vector sample files hoping that it will sink
in this time.

Post a reply to this message

Bill DeWitt wrote:

>    Actually just setting up the mirrors has been fairly simple. Making them
>move with the camera as it pans and sweeps while keeping the mirrors pointed
>at the look_at point is the hard part...

        When I want an object to move with the camera, I don't use
the look_at command.  Instead, I locate the camera at the origin <0, 0, 0>
Set the direction vector to <0, 0, 1> and then use rotations and
translation to point the camera.  I then use the same rotations and
translations on the object(s) that move with the camera.  Warning:
attempting to use a transform on a camera generates odd error
messages, so instead write a transform macro and call it from both
the camera and the object that is to move with the camera.

 #macro camera_trans()
   rotate x*7 // or whatever
   rotate y*18
   translate <0, 1, -10>
 #end

 camera {
   location 0
   direction <0, 1, 1>
   right <(640 / 480), 0, 0>
   up <0, 1, 0>
   camera_trans()
 }

 object {
   whatever
   no_shadow  // kind of handy
   translate <0, 0, 1> // so it is a little in front of the camera
   camera_trans()
 }


Pete

Post a reply to this message

"Pete" <Pet### [at] nymaliasnet> wrote in message
news:111### [at] nymaliasnet...
> Bill DeWitt wrote:
>
> >    Actually just setting up the mirrors has been fairly simple. Making
them
> >move with the camera as it pans and sweeps while keeping the mirrors
pointed
> >at the look_at point is the hard part...
>
>         When I want an object to move with the camera, I don't use
> the look_at command.

    Right, but I have to tilt the mirrors to converge on the look_at point.
If it is further away or closer, the mirror has to angle differently.

Post a reply to this message

Mirrors are slow to use if you want animation within a reasonable time (boy,
that varies person to person).
I too had made a stereo camera using the angled mirrors way, just a little
angle trickery involved.  No, nothing mathematical from me.  I found how the
subject of the view is next to impossible to get a good normal scene into
the camera.
I'm rushing through these news messages because my access was gone more than
3 days and it's catching up time so sorry if I rehash this point.
Seems Bill has come to the same conclusion.
If I remember right it was faster to render two times rather than a single
4-mirror way, for a simple scene anyhow.
That SAB POV looked great to me.  When will it be incorporated into
"official" POV-Ray? :-)
Thanks for that posting Stefan, expect e-mail as soon as I figure out where
the address is.

Bob H.

Post a reply to this message