[Insight-users] Re: problem with iterative closest point registration

[Luis Ibanez]

Hi Alireza,

If you only have four points, you could rather use the class:

         itkLandmarkBasedTransformInitializer


if you do not have the correspondence between the points, well...
this methods is so fast with only four points that you could
actually try all the combinations and settle for the one with
the smallest final squared errors.


You could also trick the Levenberg Marquardt method by providing
all the points twice. The solution should still be the same.



    Regards,


       Luis



-----------------------
Alireza Salamy wrote:
> Hi Luis,
> I try to implement iterative closestpoint2 with two pointsets which are 
> consisted of four points on each sets,but I face a problem which is:
>  
> *vnl_least_squares_function: WARNING: unknowns(6) > residuals(4)
> vnl_levenberg_marquardt: Number of unknowns(6) greater than number of 
> data (4)
> Solution = [0, 0, 0, 0, 0, 0]*
> ** 
> Is there any way to remove this problem?Shall I just change the 
> optimizer since I don't have any more points? if yes,Which optimizer do 
> you recommend me to use instead of
> itkLevenbergMarquardtOptimizer?
> It is so important for me.
> Thanks in advance for your help
> Alireza
> 
> 
> ------------------------------------------------------------------------
> Choose the right car based on your needs. Check out Yahoo! Autos new Car 
> Finder tool. 
> <http://us.rd.yahoo.com/evt=48518/*http://autos.yahoo.com/carfinder/;_ylc=X3oDMTE3NWsyMDd2BF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDY2FyLWZpbmRlcg-- 
>  >