[IGSTK-Users] Reg Tracking component

Tue Oct 26 15:14:33 EDT 2010

Hi Ziv,
You are right. The tracker would be directly connect to the system in my
application, so I don't need to broadcast the frames.
Moreover Polaris spectra that I am using uses serial interface, so
 OpenIGTLink is probably an unnecessary layer.

If you could give sources to your library, I could start getting my hands
dirty :)
Also, I have not looked deep into IGSTK code, but apart from ITK
versor/vector, vtkMatrix4x4 ,
I also found references to that IGSTK uses command/observer pattern like ITK
?
Does it use ITK itkCommand.h or implements its own ?
How deeply intertwined are IGSTK / ITK ? If its only a few classes like
igstkTransform, I can implements
my own, but if its a lot of would then I might have to skip this .

Thanks,
Somi

On Tue, Oct 26, 2010 at 12:50 PM, Ziv Yaniv <zivy at isis.georgetown.edu>wrote:

>  Hi Somi,
>
> OpenIGTLink is probably not what you are looking for. Here's how it works:
> OpenIGTLink is a message passing protocol (
> http://www.na-mic.org/Wiki/index.php/OpenIGTLink) that among others
> supports passing of transforms.
> Slicer3D (www.slicer.org) uses this protocol for tracking in image guided
> interventions. The way they do it is through IGSTK+OpenIGTLink over TCP/IP
> sockets. IGSTK provides a standalone program that does the tracking and
> transmits the OpenIGTLink messages through sockets to the Slicer
> application. You can run the two programs on the same computer or on two
> separate computers connected via ethernet.
>
> I am pretty sure this isn't what you wanted.
>
> With regard to the multi-threading library that I mentioned, this is
> original ITK code with minor modifications, mainly removal of code. Given
> that this is a modification for my uses:
> a. The license is the same as for ITK (free for any use, no strings
> attached).
> b. There is no support for this specific library as I created it for my
> uses. Given that it is 99% original ITK you can ask questions on the ITK
> list for any issues.
> c. It is minimally tested as I am relying on the original ITK testing.
> Hopefully my modifications are minor enough that I have not introduced bugs
> into the code. You can always compare its behavior to that of the official
> ITK release, just change the library to which you link, your code remains
> the same as the interface did not change. If the behavior is the same then
> my modifications did not introduce bugs that aren't already in the original
> code ;).
>
>                                hope this clarified things
>                                                     Ziv
>
>
>
> On 10/26/2010 4:31 PM, somi wrote:
>
> Thanks for the reply Patrick, Andinet, ziv,
>
>  Ziv,
> We have an application with custom features,display and tracking code
> (currently supports Polaris only) . We want to separate the tracking code
> from this application,
> so that user may be able develop and use their own trackers and are not
> restricted to Polaris.  We can implement custom tracking classes for various
> trackers,
>  but tracking is not our core competence and  we would like a toolkit which
> is widely used, robustly tested, supports major trackers and well
> documented, so that its easy for users to debug.
>
>  So I decided on using IGSTK. However if I use IGSTK I would have to
> provide ITK/VTK/IGSTK libraries with my application, even though I am using
> only
> few ITK/VTK objects ( igstk::Transform for itk , any VTK object ? ) I
> could implement my own Transform structure as you suggested. However to use
> your MT-library, there may be some issues:
>
>  a) Copyright ?? ( our software would be sold commercially in future )
> b) How much support is available for it ??
> c) Has it been rigorously tested or still in development ??
>
>
>  I also came across OpenIGTlink @
> http://www.na-mic.org/Wiki/index.php/OpenIGTLink
> I am not very clear about relation between IGSTK and OpenIGTLink ? does
> IGSTK use OpenIGTLink protocol ?
> or OpenIGTLink and IGSTK are completely different ?
>
>  Regards,
> Somi
>
> On Tue, Oct 26, 2010 at 1:54 PM, Patrick Cheng <cheng at isis.georgetown.edu>wrote:
>
>> Hi Somi,
>>
>> IGSTK is build on top of ITK/VTK. It is not possible to isolate the
>> tracking component as the way you described.
>>
>> To decouple your own application with IGSTK, you can try use the
>> OpenIGTLink library.
>> http://public.kitware.com/IGSTKWIKI/index.php/Using_OpenIGTLink_in_IGSTK
>>
>> You can start with the example application in IGSTK:
>> OpenIGTLinkTrackerBroadcasting
>>
>> Patrick
>>
>>
>> On 10/26/2010 10:42 AM, somi wrote:
>>
>>>  Hi,
>>> Do the Tracking component of IGSTK depend on ITK/VTK ?
>>>
>>> If not is it possible to build IGSTK only for tracking with no
>>> dependencies
>>> on ITK/VTK toolkits ?
>>>
>>> I am trying to integrate the tracking component of IGSTK in my
>>> application,
>>> I have my own Display/ Image processing etc code.
>>>
>>> Thanks,
>>> Somi
>>>
>>>
>>>
>>>   _______________________________________________
>>> 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
>>>
>>
>
> _______________________________________________
> 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
>
>
>
> --
> 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: +1-202-6877286
> Fax: +1-202-784-3479
> email: zivy at isis.georgetown.edu
> web: http://isiswiki.georgetown.edu/zivy/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/igstk-users/attachments/20101026/aa77cf46/attachment.html>