Difference between revisions of "2008 Summer Project Week:PerformanceTuningFiducials"
Line 33: | Line 33: | ||
''ctest -R Slicer3FiducialsTest2'' or ''source ../Slicer3/Libs/MRML/Testing/testFiducialAdd.tcl'' | ''ctest -R Slicer3FiducialsTest2'' or ''source ../Slicer3/Libs/MRML/Testing/testFiducialAdd.tcl'' | ||
− | Our plan for the project week is to first collaborate on debugging, tracking down the causes of segmentation faults during rendering with more efficient widget management algorithms, and using the Event Broker to track cascades of events.<br> | + | Our plan for the project week is to first collaborate on debugging, tracking down the causes of segmentation faults during rendering with more efficient widget management algorithms, and using the Event Broker to track cascades of events. A secondary bottleneck is in the vtkKWMultiColumnList, as updating it as fiducials are added takes a long time.<br> |
The goal is to avoid having to remove all fiducials from the scene and deleting them before readding them to reflect any changes in the MRML node.<br> | The goal is to avoid having to remove all fiducials from the scene and deleting them before readding them to reflect any changes in the MRML node.<br> | ||
If time permits, a redesign of the fiducials frame work will be sketched out, to take full advantage of the vtk widgets in VTK5.2. | If time permits, a redesign of the fiducials frame work will be sketched out, to take full advantage of the vtk widgets in VTK5.2. | ||
Line 49: | Line 49: | ||
</div> | </div> | ||
− | |||
===References=== | ===References=== | ||
* | * |
Revision as of 20:10, 12 June 2008
Home < 2008 Summer Project Week:PerformanceTuningFiducials
Key Investigators
- BWH: Nicole Aucoin
- Isomics: Steve Pieper, Alex Yarmarkovich
- Kitware: Will Schroeder
Objective
Increase responsiveness of fiducials in Slicer.
Approach, Plan
Our approach is to clear up issues with existing strategies, and then tune the fiducials to be efficient when there are 200 in the scene at once.
ctest -R Slicer3FiducialsTest2 or source ../Slicer3/Libs/MRML/Testing/testFiducialAdd.tcl
Our plan for the project week is to first collaborate on debugging, tracking down the causes of segmentation faults during rendering with more efficient widget management algorithms, and using the Event Broker to track cascades of events. A secondary bottleneck is in the vtkKWMultiColumnList, as updating it as fiducials are added takes a long time.
The goal is to avoid having to remove all fiducials from the scene and deleting them before readding them to reflect any changes in the MRML node.
If time permits, a redesign of the fiducials frame work will be sketched out, to take full advantage of the vtk widgets in VTK5.2.
Progress
TBD...