[Insight-developers] Bug 8990 - itk::FastMarchingImageFilter and initial trial points

[Dan Mueller]

Hi Arnaud,

(Sorry for the late reply, just going through all my email after vacation).

My expectation is the user specified trial value stays, otherwise
there is no/little point in the user providing trial points. In this
sense, "user-specified trial points" are more like alive points than
normal trial points.

HTH

Cheers, Dan

On 19 January 2011 18:20, Arnaud GELAS <arnaud_gelas at hms.harvard.edu> wrote:
> Luca, Dan,
>
> I was looking at the bug 8990
>
> http://public.kitware.com/Bug/view.php?id=8990
>
> and at this thread
>
> http://www.itk.org/pipermail/insight-users/2009-May/030282.html
>
> In the case where one trial point is provided by users, and it happens that
> the corresponding value is larger than the calculated one (front coming from
> another trial points). With the provided patch (actual implementation), the
> value at this trial point is never updated. (Note you discussed that point,
> but you did not address it by that time).
>
> Should this point be updated? or should we keep the one provided by the
> user?
>
> Thanks,
> Arnaud