[Paraview] PCL Plugin
Eric Younkin - NOAA Federal
eric.g.younkin at noaa.gov
Mon Jul 13 13:32:39 EDT 2015
Casey,
I don't mean to take up so much of your time, but I've been reviewing the
cmake configuration and I found that there were a number of HDF5 tests that
failed (see below). Is that significant?
*Found PythonInterp: C:/Program Files/Pydro64/python.exe (found suitable
version "2.7.3", minimum required is "2") *
*Found ZLIB:
C:/Users/eric.g.younkin/Documents/PCL_ParaView/flann-1.8.4/hdf5/lib/zlib.lib
(found version "1.2.8") *
*HDF5: SOVERSION: 8.0.2*
*HDF5: Performing Other Test STDC_HEADERS - Success*
*HDF5: Performing CXX Test OLD_HEADER_FILENAME - Failed*
*HDF5: Performing CXX Test H5_NO_NAMESPACE - Failed*
*HDF5: Performing CXX Test H5_NO_STD - Failed*
*HDF5: Performing CXX Test BOOL_NOTDEFINED - Failed*
*HDF5: Performing CXX Test NO_STATIC_CAST - Failed*
*HDF5: Checking for InitOnceExecuteOnce:*
*HDF5: Performing Test InitOnceExecuteOnce - Success*
*HDF5: Performing Other Test INLINE_TEST_inline - Failed*
*HDF5: Performing Other Test INLINE_TEST___inline__ - Failed*
*HDF5: Performing Other Test INLINE_TEST___inline - Success*
*HDF5: checking IF converting from long double to integers is accurate...
yes*
*HDF5: checking IF accurately converting from integers to long double...
yes*
*HDF5: Checking IF accurately converting unsigned long to float values...
yes*
*HDF5: Checking IF accurately roundup converting floating-point to unsigned
long long values... yes*
*HDF5: Checking IF right maximum converting floating-point to unsigned long
long values... yes*
*HDF5: Checking IF correctly converting long double to unsigned int
values... yes*
*HDF5: Checking IF overflows normally converting floating-point to integer
values... yes*
*HDF5: Checking IF correctly converting long double to (unsigned) long long
values... yes*
*HDF5: Checking IF correctly converting (unsigned) long long to long double
values... yes*
*HDF5: Checking IF alignment restrictions are strictly enforced... yes*
*HDF5: Filter ZLIB is ON*
Also, one of the errors I saw upon rebuilding:
*error C1041: cannot open program database
'c:\users\eric.g.younkin\documents\pcl_paraview\paraview-v4.3.1-source\build\plugins\sciberquesttoolkit\paraviewplugin\sciberquesttoolkit.dir\debug\vc120.pdb';
if multiple CL.EXE write to the same .PDB file, please use /FS
C:\Users\eric.g.younkin\Documents\PCL_ParaView\ParaView-v4.3.1-source\build\Plugins\SciberQuestToolKit\ParaViewPlugin\moc_pqSQFieldTracerImplementation.cpp)*
Led me to the moc_pqSQFieldTracerImplementation.cpp which had a comment at
the top stating
*"This file was generated using the moc from 5.5.0. It cannot be used with
the include files from this version of Qt. (The moc has changed too much.)"*
So I will try with Qt5.4.1 instead.
Thanks,
Eric
On Mon, Jul 13, 2015 at 10:04 AM, Eric Younkin - NOAA Federal <
eric.g.younkin at noaa.gov> wrote:
> Casey,
>
> Thanks. This is just building the latest vanilla Paraview from source.
>
> V/R,
> Eric
>
> On Mon, Jul 13, 2015 at 9:48 AM, Casey Goodlett <
> casey.goodlett at kitware.com> wrote:
>
>> Eric,
>>
>> I am not familiar with this problem.
>>
>> The code for the PCL plugin is now hosted at github here
>> https://github.com/Kitware/PCLPlugin
>>
>> Best Regards,
>>
>> On Fri, Jul 10, 2015 at 8:32 PM, Eric Younkin - NOAA Federal <
>> eric.g.younkin at noaa.gov> wrote:
>>
>>> Casey,
>>>
>>> When building Paraview 4.3.1 from source with QT5.5 in Visual Studio
>>> 2013, I get 3 failed out of 569. I pasted in the relevant parts of the VS
>>> output log below (full log too big to attach):
>>>
>>> 567> Generating Code...
>>> 567> Creating library
>>> C:/Users/eric.g.younkin/Documents/PCL_ParaView/ParaView-v4.3.1-source/build/lib/Debug/MobileRemoteControl.lib
>>> and object
>>> C:/Users/eric.g.younkin/Documents/PCL_ParaView/ParaView-v4.3.1-source/build/lib/Debug/MobileRemoteControl.exp
>>> 567> MobileRemoteControl.vcxproj ->
>>> C:\Users\eric.g.younkin\Documents\PCL_ParaView\ParaView-v4.3.1-source\build\bin\Debug\MobileRemoteControl.dll
>>> 568>C:\Program Files
>>> (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\Microsoft.CppCommon.targets(607,5):
>>> error MSB6006: "link.exe" exited with code 1120.
>>> 568>MSVCRTD.lib(crtexew.obj) : error LNK2019: unresolved external symbol
>>> WinMain referenced in function __tmainCRTStartup
>>> 568>C:\Users\eric.g.younkin\Documents\PCL_ParaView\ParaView-v4.3.1-source\build\bin\Debug\paraview.exe
>>> : fatal error LNK1120: 1 unresolved externals
>>> 569>------ Build started: Project: ALL_BUILD, Configuration: Debug x64
>>> ------
>>> 569> Building Custom Rule
>>> C:/Users/eric.g.younkin/Documents/PCL_ParaView/ParaView-v4.3.1-source/CMakeLists.txt
>>> 569> CMake does not need to re-run because
>>> C:\Users\eric.g.younkin\Documents\PCL_ParaView\ParaView-v4.3.1-source\build\CMakeFiles\generate.stamp
>>> is up-to-date.
>>> ========== Build: 566 succeeded, 3 failed, 0 up-to-date, 0 skipped
>>> ==========
>>>
>>> Doing some Google searches seem to indicate that in order to clear the
>>> LNK2019 error, I would need to change the project type
>>> <http://stackoverflow.com/questions/6626397/error-lnk2019-unresolved-external-symbol-winmain16-referenced-in-function>.
>>> And that in order to do that, I would need to change the configuration type from
>>> utility to something else
>>> <http://gotoanswer.com/?q=No+Linker+option+in+Visual+Studio+Project+Properties>
>>> .
>>>
>>> I suspect that I am just missing something obvious, but I'm not sure
>>> what it is. Can you offer any advice?
>>>
>>> Thanks,
>>> Eric
>>>
>>> On Fri, Jul 10, 2015 at 12:03 PM, Casey Goodlett <
>>> casey.goodlett at kitware.com> wrote:
>>>
>>>> You should compile against the VTK of paraview. I have not tested Qt
>>>> 5.5.
>>>>
>>>> I have not recently tested building VeloView against paraview outside
>>>> of the VeloView superbuild. There may be unexpected problems.
>>>>
>>>> On Fri, Jul 10, 2015 at 11:49 AM, Eric Younkin - NOAA Federal <
>>>> eric.g.younkin at noaa.gov> wrote:
>>>>
>>>>> Thanks Casey. I'm reading in the Veloview Developer guide (
>>>>> https://github.com/Kitware/VeloView/tree/master/Documentation) that
>>>>> Veloview requires VTK6 and Qt 4.8. I'm currently compiling Paraview from
>>>>> source using Qt5.5 and VTK libraries that came with Paraview source. Do I
>>>>> need to download VTK6 and Qt4.8 and use those? I've had some issues with
>>>>> VS2013 and compiling with Qt4.
>>>>>
>>>>> Also, I'm currently getting the following repeating error upon
>>>>> configuring in CMake:
>>>>>
>>>>> CMake Warning (dev) in
>>>>> ThirdParty/QtTesting/vtkqttesting/CMakeLists.txt:
>>>>> Policy CMP0020 is not set: Automatically link Qt executables to qtmain
>>>>> target on Windows. Run "cmake --help-policy CMP0020" for policy
>>>>> details.
>>>>> Use the cmake_policy command to set the policy and suppress this
>>>>> warning.
>>>>> This warning is for project developers. Use -Wno-dev to suppress it.
>>>>>
>>>>> I'm seeing some references on the VTK users group (
>>>>> http://public.kitware.com/pipermail/vtkusers/2013-July/080543.html)
>>>>> that seem to indicate you can ignore them. Any reason I shouldn't continue
>>>>> with Qt5.5?
>>>>>
>>>>> Thanks,
>>>>> Eric
>>>>>
>>>>> On Fri, Jul 10, 2015 at 10:43 AM, Casey Goodlett <
>>>>> casey.goodlett at kitware.com> wrote:
>>>>>
>>>>>> Eric,
>>>>>>
>>>>>> That paper describes the approach for filter wrapping. You will also
>>>>>> need to compile VeloView against your paraview build and then include the
>>>>>> VeloView plugin alongside the PCL plugin.
>>>>>>
>>>>>> Hope that helps.
>>>>>>
>>>>>> On Thu, Jul 9, 2015 at 3:00 PM, Eric Younkin - NOAA Federal <
>>>>>> eric.g.younkin at noaa.gov> wrote:
>>>>>>
>>>>>>> I should also mention that I have been reading this paper
>>>>>>> <http://www.paraview.org/Wiki/images/1/1e/Paper.pdf> and a few
>>>>>>> other sources. I think I have a basic idea of the framework.
>>>>>>>
>>>>>>> On Thu, Jul 9, 2015 at 11:33 AM, Eric Younkin - NOAA Federal <
>>>>>>> eric.g.younkin at noaa.gov> wrote:
>>>>>>>
>>>>>>>> Thanks Casey. For someone with only moderate programming
>>>>>>>> experience, how difficult would you say creating those additional wrappers
>>>>>>>> and building a VLP-16 paraview plugin would be? Do you happen to have any
>>>>>>>> resources or guides for doing so?
>>>>>>>>
>>>>>>>> I will continue to look at building from source and check out
>>>>>>>> Project Tango. Thanks for the link.
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Eric
>>>>>>>>
>>>>>>>> On Thu, Jul 9, 2015 at 11:14 AM, Casey Goodlett <
>>>>>>>> casey.goodlett at kitware.com> wrote:
>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Thu, Jul 9, 2015 at 10:53 AM, Eric Younkin - NOAA Federal <
>>>>>>>>> eric.g.younkin at noaa.gov> wrote:
>>>>>>>>>
>>>>>>>>>> Casey,
>>>>>>>>>>
>>>>>>>>>> I am mostly interested in the feature extraction routines and
>>>>>>>>>> Paraview integration with the Velodyne VLP-16. I'd like to experiment with
>>>>>>>>>> PCL algorithms to try and pick out fixed objects such as buoys or piers
>>>>>>>>>> from a VLP-16 mounted on our research vessel as we travel up the coast. I
>>>>>>>>>> think most of my errors are from attempting to build using Visual Studio
>>>>>>>>>> 2013 pro and older Qt/PCL versions. If I can use the most recent versions,
>>>>>>>>>> that might help things.
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Sounds like a cool project. You may need to add some additional
>>>>>>>>> wrappers for feature extraction if you are looking for things beyond the
>>>>>>>>> sample consensus models that we have. I dont think any of the point
>>>>>>>>> feature histogram utilities are currently wrapped.
>>>>>>>>>
>>>>>>>>> Windows builds of the whole stack are a bit tricky. Feel free to
>>>>>>>>> share your error messages, and I'll help where I can.
>>>>>>>>>
>>>>>>>>> The plugin does not include support for the VLP-16 in its current
>>>>>>>>> state. We have another project VeloView
>>>>>>>>> <https://github.com/Kitware/VeloView> that builds a paraview
>>>>>>>>> plugin for VLP-16. I have used this in combination with the PCL plugin in
>>>>>>>>> the past, but the build is not straightforward at this time.
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> The binaries you linked to here still require installation of
>>>>>>>>>> Paraview and PCL correct?
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>> This binary is ParaView with the pcl plugin built and packaged.
>>>>>>>>> If you download this it includes paraview and PCL.
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Casey B Goodlett, PhD
>>>>>>>>> Technical Leader
>>>>>>>>> Kitware, Inc. - North Carolina Office
>>>>>>>>> http://www.kitware.com
>>>>>>>>> (919) 969-6990 x310
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> Eric Younkin LTJG/NOAA
>>>>>>>> HSTP Liaison - East
>>>>>>>> Marine Operations Center - Atlantic
>>>>>>>> 439 W York St
>>>>>>>> Norfolk, VA 23510
>>>>>>>> Cell: 828-331-8197
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> Eric Younkin LTJG/NOAA
>>>>>>> HSTP Liaison - East
>>>>>>> Marine Operations Center - Atlantic
>>>>>>> 439 W York St
>>>>>>> Norfolk, VA 23510
>>>>>>> Cell: 828-331-8197
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Casey B Goodlett, PhD
>>>>>> Technical Leader
>>>>>> Kitware, Inc. - North Carolina Office
>>>>>> http://www.kitware.com
>>>>>> (919) 969-6990 x310
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Eric Younkin LTJG/NOAA
>>>>> HSTP Liaison - East
>>>>> Marine Operations Center - Atlantic
>>>>> 439 W York St
>>>>> Norfolk, VA 23510
>>>>> Cell: 828-331-8197
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Casey B Goodlett, PhD
>>>> Technical Leader
>>>> Kitware, Inc. - North Carolina Office
>>>> http://www.kitware.com
>>>> (919) 969-6990 x310
>>>>
>>>
>>>
>>>
>>> --
>>> Eric Younkin LTJG/NOAA
>>> HSTP Liaison - East
>>> Marine Operations Center - Atlantic
>>> 439 W York St
>>> Norfolk, VA 23510
>>> Cell: 828-331-8197
>>>
>>
>>
>>
>> --
>> Casey B Goodlett, PhD
>> Technical Leader
>> Kitware, Inc. - North Carolina Office
>> http://www.kitware.com
>> (919) 969-6990 x310
>>
>
>
>
> --
> Eric Younkin LTJG/NOAA
> HSTP Liaison - East
> Marine Operations Center - Atlantic
> 439 W York St
> Norfolk, VA 23510
> Cell: 828-331-8197
>
--
Eric Younkin LTJG/NOAA
HSTP Liaison - East
Marine Operations Center - Atlantic
439 W York St
Norfolk, VA 23510
Cell: 828-331-8197
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20150713/1bb7ac84/attachment.html>
More information about the ParaView
mailing list