[osg-users] Transparent but occluding phantom objects for AR application

2010-07-09 Thread Simon Julier

Dear All,

I'm sure this has been answered before, and I'm merely missing the  
magic combination of keywords in my various searches.


I'm developing a video-see-through AR application which will show the  
user the contents of holes in the ground. To get the correct depth  
cues, the user should be able to see the contents of the hole only  
when they look down into it. From what I can tell, the right way to  
achieve this is to create some phantom objects which are transparent  
but occlude objects placed behind them. These phantom objects should  
be realised by writing them into the depth buffer only.


Is that correct? If so, could anybody point me towards a suitable  
reference?


Many thanks!

Cheers,

Simon
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] Transparent but occluding phantom objects for AR application

2010-07-09 Thread Farshid Lashkari
Hi Simon,

On Thu, Jul 8, 2010 at 11:09 PM, Simon Julier s.jul...@cs.ucl.ac.uk wrote:

 I'm developing a video-see-through AR application which will show the user
 the contents of holes in the ground. To get the correct depth cues, the user
 should be able to see the contents of the hole only when they look down into
 it. From what I can tell, the right way to achieve this is to create some
 phantom objects which are transparent but occlude objects placed behind
 them. These phantom objects should be realised by writing them into the
 depth buffer only.

 Is that correct? If so, could anybody point me towards a suitable
 reference?


You can use the osg::ColorMask state attribute to prevent the object from
writing to the color buffer. I've used this technique in my
video-see-through apps.

Cheers,
Farshid
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] [vpb] .tif and .tfw

2010-07-09 Thread lucie lemonnier
Hello Pierre,
I send you in attachment what I get in my console with OSG_NOTIFY_LEVEL = 
INFO when I run osgdem --geocentric -d data.dem -t data.tif -l 3 -o data.osg 
and when I run osgviewer data.osg.
Otherwise, I use osg-2.8.2, VPB 0.9.10 and gdal 1.5.
I will try to use Boston data.


Thank you!

Cheers,
lucie

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29858#29858




Attachments: 
http://forum.openscenegraph.org//files/run_osgviewer_830.txt
http://forum.openscenegraph.org//files/run_osgdem_171.txt


___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] osganimationsolid example

2010-07-09 Thread Gianni Ambrosio
Thank you Jean-Sébastien for the detailed explanation.
I need to move objects related to cars and their positions/rotations are 
already calculated by a solver. So I think I will try the callback approach.

Thanks again,
Gianni

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29859#29859





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] Texture problem on 64-bit PC

2010-07-09 Thread Andreas Ekstrand

Hi,

I have similar problems with an ATI Radeon HD 5850 in Windows 7 64bit. 
Textures are corrupted, it seems like the MIP levels are mixed together 
somehow, so that different textures are shown and blended together 
depending on the viewing distance and angle. I have tried different 
versions of the Catalyst drivers but to no avail. The phenomenon doesn't 
appear with all models but it feels like it's an ATI driver issue 
anyway, triggered by something special in the failing models, but I 
haven't figured out what just yet.


Does anyone else have the same problem or know how to solve it?

Regards,
Andreas


On 2010-07-08 17:48, Gordon Tomlinson wrote:

I would have to disagree and say that it is likely to be a hardware\driver
issue

I have been running OSG on 64 with Nvidia for 3 years and had no issues

There have bee quite a few posts recently about continued issues with ATI
catalyst drivers and their broken Opengl Support, the suggestions from the
post is to use catalyst  10.4 drivers



__
Gordon Tomlinson

gor...@gordontomlinson.com IM: gordon3db...@3dscenegraph.com
www.vis-sim.com
www.gordontomlinson.com
www.PhotographyByGordon.com

__

-Original Message-
From: osg-users-boun...@lists.openscenegraph.org
[mailto:osg-users-boun...@lists.openscenegraph.org] On Behalf Of Werner
Stoop
Sent: Thursday, July 08, 2010 4:29 PM
To: osg-users@lists.openscenegraph.org
Subject: [osg-users] Texture problem on 64-bit PC

Hi,

I recently received a new 64-bit machine at work. When I now use osgviewer
to the textures on some of our models seem smeared where they displayed
correctly on my older 32-bit PC.

Other models do display correctly, so I don't think it's a hardware issue.

Our models are in the form of IVE files, and some of them are generated by
(an older version of) Virtual Planet Builder.

For what it's worth, my PC runs Fedora 12 and has an ATI Radeon HD 3450
graphics card. I also had Fedora 12 on my previous PC, but it had a NVidia
card.

I would post an image, but I'm a newbie, so I'm not allowed to post URLs

Has anyone had similar problems? what causes it? How can I resolve it or fix
the models?

Thank you!
Werner[/i]

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29837#29837





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

   


___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] [vpb] .tif and .tfw

2010-07-09 Thread lucie lemonnier
Hi,
I tested the boston example but I'm not sure that the projection has worked 
well.
I run osgdem.exe --geocentric -d boston-dem-downtown-5m-utm19n-meters.tif -t 
boston-10m-utm19n-meters.tif -l 5 -o test2.osg.
I send what I obtain in attachment.
Can you tell me if it worked well?


Thank you!

Cheers,
lucie

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29863#29863




Attachments: 
http://forum.openscenegraph.org//files/boston_184.jpg
http://forum.openscenegraph.org//files/boston_171.rar


___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


[osg-users] osgVolume volume rendering ?

2010-07-09 Thread Emmanuel Roche
Hi everyone,

I'm just starting to play with the osgVolume sample and I have a very
beginner question about it:

- Can you actually have any kind of volumetric rendering with this ???

I perform many tests with various arguments,  but each time, all I get is a
simple 2D image sequence of all the provided 2D images: so it's basically
playinig just like a movie... Why would we have complex Raytracing  shader
techniques just to render an image sequence ? I must be missing something
here.


Also, note that I actually managed to get a 3D rendering specifying zSize,
ySize and xSize, but again the 3D result  is merly the projection of a
single 2D image (an the image is dynamically changing...) I know the
osgVolume still needs more development, but is that really all I can expect
from it for the moment ??

Also, I could not use DICOM images directly: I got errors from the dcmtk
library, so I first used image magick to convert my dcm images to png
images, could it be I lost usefull 3D volumetric rendering info with that
step ?

And last but not least, those tests were performed on Windows, with OSG
2.9.6. Any known limitation due to this ?


In fact, if anyone as any screenshot I would be interested ! :-)

Cheers,
Manu.
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] [vpb] .tif and .tfw

2010-07-09 Thread Pierre Bourdin (gmail)
Hello Lucie,

When looking at your log I see that the tiff texture is reprojected...

started DataSet::createDestination(3)
reprojecting to file temporaryfile_013c07_0100_deme.tif
reprojecting to file
temporaryfile_S4_06032_5215_20061018_m20_1_utm20.tif
Time for after_reproject 8.389910

as recommended by Shayne, you should re-project your data, first it will
save you some processing time and secondly it will allow you to detect
projection problems more easily...


To avoid aliasing artifacts when accessing data, you can also use
gdaladdo to store the mip maps in the .tiff, like explained here :
http://www.openscenegraph.org/projects/osg/wiki/Support/UserGuides/osgdem
(but it won't fix your projection trouble)

Otherwise I believe there was some problems with GDAL 1.5 at least with
some release ;  particularly the 1.5.0 was buggy...

If using known to work data, you still got strange result, try to update
gdal lib...

I've downloaded and tested on my machine (Debian 64bits - osg 2.9.7)
your Boston terrain and it looks normal... but the files are probably
not reprojected...

You could also try to look at your data in a GIS tool like Quantum GIS
http://www.qgis.org/

Just to see if everything is looking good ?


Pierre.

Le vendredi 09 juillet 2010 à 07:45 +, lucie lemonnier a écrit :
 Hello Pierre,
 I send you in attachment what I get in my console with OSG_NOTIFY_LEVEL = 
 INFO when I run osgdem --geocentric -d data.dem -t data.tif -l 3 -o 
 data.osg and when I run osgviewer data.osg.
 Otherwise, I use osg-2.8.2, VPB 0.9.10 and gdal 1.5.
 I will try to use Boston data.
 
 
 Thank you!
 
 Cheers,
 lucie
 
 --
 Read this topic online here:
 http://forum.openscenegraph.org/viewtopic.php?p=29858#29858
 
 
 
 
 Attachments: 
 http://forum.openscenegraph.org//files/run_osgviewer_830.txt
 http://forum.openscenegraph.org//files/run_osgdem_171.txt
 
 
 ___
 osg-users mailing list
 osg-users@lists.openscenegraph.org
 http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org

-- 

Pierre BOURDIN
I.M.E.R.I.R.
Av. Pascot BP 90443
66004 PERPIGNAN
tél: 04 68 56 84 95
fax: 04 68 55 03 86
email: bour...@imerir.com


___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] FBX Plugin: animation speed not as expected...

2010-07-09 Thread Alessandro Terenzi
I've downloaded the latest source code of the FBX plugin and osgAnimation
from svn but, I'm still experiencing different playback speed with respect
to QuickTime moreover I have also problems with the animations themselves.

As I said I'm using the latest code for FBX plugin and osgAnimation, on the
other hand I'm using OSG 2.9.6 for all other plugins and libraries, so my
question is: beyond the FBX plugin itself and osgAnimation what else could
cause the problems with animations I'm experiencing? (ie not only speed, but
also wrong/bad animation playback in general)

By the way, I tried with both QuickTime and the ViewScene example that comes
with the FBX SDK, they both play every model correctly. Talking about test
models, I tryed:

humanoid.fbx: model messed up during animation
the 3-bouncing-balls model (sent to Michael some days ago): animation played
at exagerated speed

and now I have two more humanoid models one of which has the same problems
of humanoid.fbx and the other seems to be missing some starting and ending
frames. I'm going to send them to Michael.

Regards.
Alessandro


On Thu, Jul 8, 2010 at 9:35 AM, Alessandro Terenzi a.tere...@gmail.comwrote:

 Ok, thanks Michael, I will double check that I'm really using the latest
 source code for both the FBX plugin and osgAnimation.

 Regards.
 Alessandro

 On Wed, Jul 7, 2010 at 6:54 PM, Michael Platings mplati...@gmail.comwrote:

 Hi Alessandro, I can't reproduce the problem so I don't know why you're
 seeing that, sorry.

 ___
 osg-users mailing list
 osg-users@lists.openscenegraph.org
 http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org



___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] osganimationsolid example

2010-07-09 Thread Gianni Ambrosio
One more question.
The solver I have generates positions and rotations for the object to move at 
runtime. So I can not set all positions and rotations when the animation 
starts. In case of using the osgAnimation library, how could I append new 
positions and rotations at runtime?

Regards
Gianni

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29869#29869





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] Qt 4.6 context menus not repainting in osgViewerQT or osgViewerQtWidget

2010-07-09 Thread James Davis
I have just upgraded to qt 4.6.3, and it has corrected the problem.  Thanks, 
Pan Liang, for the suggestion.

James

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29870#29870





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] osganimationsolid example

2010-07-09 Thread Jean-Sébastien Guay

Hi gianni,


The solver I have generates positions and rotations for the object to move at 
runtime. So I can not set all positions and rotations when the animation 
starts. In case of using the osgAnimation library, how could I append new 
positions and rotations at runtime?


Err, I wonder, if you have something else calculating positions and 
rotations at runtime, why do you want to use osgAnimation or a callback 
at all? Just set your graphical object's transform to the new values and 
you're done...


osgAnimation or callbacks will be useful if you have a sequence of 
precalculated positions/orientations for each frame of animation, and 
then you want at runtime to move through the sequence and set the 
current position/orientation. Or if you want to calculate the current 
position yourself on the fly. But if you have a separate system 
calculating the position/orientation, then just set the graphical 
object's transform to that new value directly.


Hope this helps,

J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
   http://www.cm-labs.com/
http://whitestar02.webhop.org/
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] DatabasePager Callback

2010-07-09 Thread Guy Volckaert
I am aware of the ReadFileCallback - but I guess the problem is how to 
distinguish between my paging terrain files from other database files loaded 
during runtime. All my files are FLTs (with the exception of a few TXP files 
for the terrain). I can't use the file extension as a filter since they are all 
the same (i.e *.flt).

Keep the ideas comming... it's all good.

Regards,

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29872#29872





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] Transparent but occluding phantom objects for AR application

2010-07-09 Thread S . Julier

Dear Farshid,

You can use the osg::ColorMask state attribute to prevent the object 
from writing to the color buffer. I've used this technique in my 
video-see-through apps.


Thanks. I tried it, but it didn't seem to work for me. The way I did it is 
as follows (r is of type osg::Node*):


 osg::StateSet* stateSet = r-getOrCreateStateSet();
 osg::ColorMask* cm = new osg::ColorMask(false, false, false, false);
 stateSet-setAttributeAndModes( cm, osg::StateAttribute::ON);

It seems to work fine for masking the bits off for each colour channel 
separately. However, when I disable all the channels, the phantom object 
does not appear to occlude objects behind it.


I also made sure that the depth test was enabled, but that didn't make any 
difference to the results.


Any thoughts?

Cheers,

Simon

___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] osganimationsolid example

2010-07-09 Thread Gianni Ambrosio
Hi Jean-Sébastien,
I would like to handle two slightly different scenarios.

1) Rotation  orientation values are loaded from file (already generated by the 
solver). Those values are just like the channels of an animation, + one of them 
is the time channel (with fixed step).

2) The solver calculates the rotations  orientations but since it is faster 
than realtime I need to store them in the same class of point 1. In fact I 
cannot update the transform directly bucause I would run the animation faster 
than realtime.

At point 1) I could set the animation values but at point 2 I have to add them 
at runtime, while the animation is running.

Maybe what you suggested is correct but it seems the callback method allows me 
to update the transform when the three is traversed by the update visitor. 
Moreover I need a way to show the animation in realtime so I need a way to 
synchronize the transform update to the time channel.

Hope this may clarify my problem

Gianni

--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29874#29874





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] osganimationsolid example

2010-07-09 Thread Jean-Sébastien Guay

Hello Gianni,


Maybe what you suggested is correct but it seems the callback method allows me 
to update the transform when the three is traversed by the update visitor. 
Moreover I need a way to show the animation in realtime so I need a way to 
synchronize the transform update to the time channel.


You're right, using a callback is appropriate if you need to update your 
positions/orientations based on a time and you get values from a file or 
from a generator that generates them faster than realtime and you cache 
the values.


As for the time, in your callback's operator(), the nodevisitor you get 
has a getFrameStamp() method that lets you get the reference time, which 
is the time since the program started running. From this you can decide 
which position/orientation values to use from your cache (and perhaps 
expire older values, or whatever). Anyways, only you will know what you 
want to do, but the information is there for you to make your decisions.


Hope this helps,

J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
   http://www.cm-labs.com/
http://whitestar02.webhop.org/
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] Transparent but occluding phantom objects for AR application

2010-07-09 Thread Farshid Lashkari
Hi Simon,

On Fri, Jul 9, 2010 at 7:38 AM, s.jul...@cs.ucl.ac.uk wrote:

 It seems to work fine for masking the bits off for each colour channel
 separately. However, when I disable all the channels, the phantom object
 does not appear to occlude objects behind it.


You will need to ensure that the phantom objects are rendered before the
objects they are supposed to occlude. You can do this by either giving the
phantom objects a very low render bin order, or the real objects a very
high render bin order. The following sample shows how to give an
osg::StateSet a very low draw order:

stateset-setRenderBinDetails( -1000, RenderBin,
osg::StateSet::USE_RENDERBIN_DETAILS );

Cheers,
Farshid
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] problem with osgUtil::IntersectVisitor and osg::LineSegment

2010-07-09 Thread Tim Larson
I stumbled on this thread when working on a similar problem that I had using 
IntersectionVisitor in Visual Studio projects.


I just wanted to confirm that I have found that the OSG examples and OSGHelp 
examples that utilized IntersectionVisitor crashed under Windows when utilizing 
the latest downloaded Visual Studio binaries.

(I tried the exaples MouseSceneIntersection, 
osgpick and osgkeyboardmouse and could not get any of them to function)

However, numerous other example apps are working in Visual Studio.


Based on the previous posters response it appears that recompiling the Source 
within Visual studio is the method to address this problem.

 
omartinaud wrote:
 Hi everybody,
 
 
 Finally, I decided to reinstall osg using the sources. I took the osg 2.8.3 
 sources and I compiled them using my version of visual studio. Now everything 
 works perfectly ! Looks like I didn't have the good binaries.
 
 Thank you all for your help !
 
 Cheers,
 Olivier


--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=29877#29877





___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] problem with osgUtil::IntersectVisitor and osg::LineSegment

2010-07-09 Thread Jean-Sébastien Guay

Hello Tim,


Based on the previous posters response it appears that recompiling the Source 
within Visual studio is the method to address this problem.


Err, no, usually just making sure you don't mix debug and release and 
that you use binaries for the same version of Visual Studio should make 
everything work. The previous poster was me, and I just confirmed that 
the OP was using VC9 (Visual Studio 2008) and the binaries they had 
downloaded were for VC8 (Visual Studio 2005).


Of course, in their case there was no other option than compiling OSG 
from source, since there were no binaries for VC9 available at the time. 
It appears there still aren't, so if you use VC9 then you'll have to 
compile from source too. But saying that the solution to the problem 
above was recompiling from source is false. The solution was to use 
binaries for the same version of Visual Studio, which in this particular 
case implied compiling from source.


It just means you have to know what you're downloading. If you see 
binaries for Visual Studio 8 and you're using Visual Studio 9, don't 
assume it will still work and then wonder why it crashes...


Which version of Visual Studio are you using? Which version of Visual 
Studio are the binaries compiled for? What library files are you linking 
to? What configuration are you compiling your project in?


J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
   http://www.cm-labs.com/
http://whitestar02.webhop.org/
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org


Re: [osg-users] problem with osgUtil::IntersectVisitor and osg::LineSegment

2010-07-09 Thread Jean-Sébastien Guay

Hi Tim,


It just means you have to know what you're downloading. If you see
binaries for Visual Studio 8 and you're using Visual Studio 9, don't
assume it will still work and then wonder why it crashes...


In fact, the download page even states it in underlined text...

- Quote -
Visual Studio users, make sure you download the binaries for your Visual 
Studio version: 8 (2005) or 9 (2008). You cannot link OSG libraries for 
VS8 into a VS9 executable or vice versa, they have to match. If you want 
to be really sure, download the source and compile OSG yourself.

-

http://www.openscenegraph.org/projects/osg/wiki/Downloads

J-S
--
__
Jean-Sebastien Guayjean-sebastien.g...@cm-labs.com
   http://www.cm-labs.com/
http://whitestar02.webhop.org/
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org