[IGSTK-Users] Augmented reality

[Ziv Yaniv]

Hi Alvaro,

Right now the igstkPerspectiveTransform is more of a place holder than a 
class that is in use. I implemented it as a data object for loading 
calibration data for perspective cameras (endoscope/microscope/X-ray).

Up to this point we have not developed a framework for augmented reality 
in IGSTK so we don't offer an "official" solution that is better than 
your use of multiple renderers. Keep in mind that the coordinate system 
tree you are creating makes sense for your application.

Adding such a framework is one of the extensions of the toolkit we will 
be looking into.

               regards
                   Ziv

-- 
Ziv Yaniv, PhD., Research Assistant Professor
Imaging Science and Information Systems (ISIS) Center
Department of Radiology
Georgetown University Medical Center
2115 Wisconsin Avenue, Suite 603
Washington, DC, 20007,

Phone: +1-202-687-7286
Fax: +1-202-784-3479
email: zivy at isis.georgetown.edu
web: http://isiswiki.georgetown.edu/zivy/

Alvaro Gómez wrote:
> 
> The idea of using the perspective transform seems to be the better way 
> but I still don´t see how to use it and I don´t find examples in the 
> IGSTK or Sandbox code.
> 
> Could you give me the basic idea of how to use the 
> igstkPerspectiveTransform or point me an example?
> 
> Thanks
> Alvaro
> 
> El día martes, 23 de junio de 2009 05:38:44 p.m.
> Patrick Cheng <cheng at isis.georgetown.edu>
> escribió:
>> Hi Alvaro,
>>
>>
>>
>> Sorry for the late reply. Your email probably fall through our eyes as 
>> we were busy preparing for the latest release.
>>
>>
>>
>> I think what you are doing is feasible. My question is why do you want 
>> to use multiple renderers?
>>
>>
>>
>> If you want to overlay object with a 2D video, you can use perspective 
>> transform to project the object on to the image 
>> (igstkPerspectiveTransform). It will be better to use a single 
>> renderer so that you have the same camera and coordinate system.
>>
>>
>>
>> Also notice that in the IGSTK coordinate system frame work, we 
>> consider View as a node in the scene graph tree. That assumes we only 
>> have one renderer in the View (one coordinate system and one camera 
>> position). Take these into consideration if you want to modify View to 
>> accommodate multiple renderers.
>>
>>
>>
>> Patrick
>>
> 
> _______________________________________________
> Powered by www.kitware.com
> 
> Visit other Kitware open-source projects at 
> http://www.kitware.com/opensource/opensource.html
> 
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/cgi-bin/mailman/listinfo/igstk-users
> 
>