[#4231] improve performance of affine groupwise registration for many small images

Please log in

State: more information
Open
Date:
2009-10-01 16:26
Priority: more information
3
Submitted By:
Torsten Rohlfing (torstenrohlfing)
Assigned To: more information
Torsten Rohlfing (torstenrohlfing)
Product: 
CMTK-core
Operating System: 
All
Component: 
Libs
Summary: more information
improve performance of affine groupwise registration for many small images

Detailed description

Affine groupwise registration uses data parallelism by image for SMP computation.

If a large number of small images are registered, the algorithm scales extremely poorly.

A second level of parallelism should be added, most importantly over the components of the cost function gradient. That will require some major changes, however, because there is currently only temporary space for one reformatted image, and the "Evaluate()" function always works on the object's m_Data[] array.

Response

No Responses Have Been Posted

Attached Files:

Name Download
No Files Currently Attached

Changes:

No Changes Have Been Made to This Item