[Insight-developers] question about itkImagePCAShapeModelEstimator

[Arnaud GELAS]

Hi guys,

I am going through the code, and I have few questions / comments

    * What's the benefit of using vnl_generalized_eigensystem vs vnl_svd
      or vnl_symmetric_eigensystem? (I may miss something here)

    * It is difficult to know in advance how many eigen vectors and
      values you would need for your application. In some particular
      case one user could provide a ratio (like 80% of the variation),
      the algorithm first compute the eigen-decomposition and just keep
      the ratio * the largest eigen values / corresponding eigen
      vectors. Are there people interested in this?


Arnaud

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20100422/aa9ffdb9/attachment.htm>