[Paraview] can vtkImplicitPlaneWidget be used in an XML-only paraview plug-in?

[Dr. Roman Grothausmann]

Dear Utkarsh,

On 12/12/14 19:13, Utkarsh Ayachit wrote:
> you want a way to clip an image dataset using an arbitrary
> clip plane (not necessarily axis aligned) and then volume render it.

Yes, if possible any kind of implicit function.

> If you use the Clip fliter, the resulting unstructured grid becomes
> too large and hence takes a lot of time to volume render. Is that
> correct?

Not only the rendering takes a long time but even the execution of the filter 
itself, and its precision on cutting voxels is not at all needed for a volume of 
around 1500x1500x1500 voxels. Basically I just want to make voxels 
inside/outside the implicit function be "invisible".
What's the best way to achieve that?

Many thanks for Your help.
Roman

>
> BTW, if you want to just clip using axis-aligned clip plane, you
> should just use the "Extract Subset" filter.
>
> Utkarsh
>
> On Thu, Dec 11, 2014 at 3:57 AM, Dr. Roman Grothausmann
> <grothausmann.roman at mh-hannover.de> wrote:
>> Dear Utkarsh,
>>
>>
>> Just yesterday I stumbled over vtkImplicitTextureCoords. Could that be used
>> to achieve my initial idea, ie to avoid the paraview clipper for clipping
>> images "roughly"? I managed to create an xml-file that loads as a
>> filter-plugin in paraview but after execution, there seems to be no texture
>> available to use for rendering. I'm not sure if using ProxyProperty,
>> ProxyGroupDomain and ProxyListDomain is doing the same for a filter instead
>> of a source as in Your original XML. Are those tags the same as
>> InputProperty, DataTypeDomain otherwise used for the input of a
>> filter-plugin?
>> If that is correct, what else could I be missing?
>>
>> Many thanks for looking into this.
>> Roman
>>
>> On 26/08/13 15:06, Utkarsh Ayachit wrote:
>>>>
>>>> Many thanks for Your reply, Your effort to create an XML-plugin and the
>>>> bug
>>>> report. Would it help to compile paraview from source with
>>>> Module_vtkImagingStencil set to ON on our own platform for testing or is
>>>> there some code adjustment of paraview necessary to allow that?
>>>
>>>
>>> No, it won't help. I did exactly that an ran into issues with
>>> vtkImageStencil and ParaView. Evidently, we hadn't tried to use that
>>> data-object in ParaView before.
>>>
>>>> I guess it would not help to specify the location of a separately
>>>> compiled vtk-5.10?
>>>
>>>
>>> You are correct, it will not help. We need to track the issue down on
>>> the ParaView side first.
>>>
>>>> Do I understand correctly that with using port_index= You specify which
>>>> output should go to which input?
>>>
>>>
>>> That is correct. If not specified, 0 is assumed.
>>>
>>>> What is calling e.g. the vtkImplicitPlaneWidget to get the interaction
>>>> widget for specifying the implicit function parameters?
>>>
>>>
>>> In this case, it's the "ProxyListDomain" associated with the
>>> "ImplicitFunction" property. The domain refers to proxies viz.
>>> ("implicit_functions", "Plane"), etc. If you look at the xml
>>> definitions for the same in ParaView
>>> (ParaViewCore/ServerManager/SMApplication/Resources/utilities.xml),
>>> you'll see that it provides a "<Hints>" section that suggests a
>>> "PropertyGroup" of type "Plane". That's what's causing the GUI to show
>>> the vtkImplicitPlaneWidget to control the Origin and Normal
>>> properties.
>>>
>>> Hope that clarifies things a bit.
>>>
>>> Utkarsh
>>>
>>
>> --
>> Dr. Roman Grothausmann
>>
>> Tomographie und Digitale Bildverarbeitung
>> Tomography and Digital Image Analysis
>>
>> Institut für Funktionelle und Angewandte Anatomie, OE 4120
>> Medizinische Hochschule Hannover
>> Carl-Neuberg-Str. 1
>> D-30625 Hannover
>>
>> Tel. +49 511 532-9574

-- 
Dr. Roman Grothausmann

Tomographie und Digitale Bildverarbeitung
Tomography and Digital Image Analysis

Institut für Funktionelle und Angewandte Anatomie, OE 4120
Medizinische Hochschule Hannover
Carl-Neuberg-Str. 1
D-30625 Hannover

Tel. +49 511 532-9574