[Insight-users] Complexity of the metric?

[motes motes]

The question was related to the MeanSquaresImageToImageMetric, which
as I understand always will be linear in the total number of voxels.



On Fri, Dec 11, 2009 at 9:44 PM, Bill Lorensen <bill.lorensen at gmail.com> wrote:
> Depends on the metric. Some statistical metrics sample a fixed number
> of samples. Look at the mutual information metrics.
>
> On Fri, Dec 11, 2009 at 3:22 PM, motes motes <mort.motes at gmail.com> wrote:
>> As I understand the metric is the most expensive part of the image
>> registration process. I assume that this follows from the fact that it
>> is linear in the number of voxels, meaning that if a volume has m*n*p
>> dimensions the number of computations of the metric is m*n*p. When
>> m=n=p is it then correct to say that the complexity of the metric is
>> O(n^3) ?
>> _____________________________________
>> Powered by www.kitware.com
>>
>> Visit other Kitware open-source projects at
>> http://www.kitware.com/opensource/opensource.html
>>
>> Kitware offers ITK Training Courses, for more information visit:
>> http://www.kitware.com/products/protraining.html
>>
>> Please keep messages on-topic and check the ITK FAQ at:
>> http://www.itk.org/Wiki/ITK_FAQ
>>
>> Follow this link to subscribe/unsubscribe:
>> http://www.itk.org/mailman/listinfo/insight-users
>>
>