[vtkusers] can we match a cropped version of a model againt the original using the vtkIterativeClosestPointTransform?

[Wes Turner]

The filter handles the selection automatically.  Your just need to set the
maximum number of points you want to use for the landmarks.  Check out
/Hybrid/Testing/Tcl/TestICPTransform.tcl
<http://public.kitware.com/cgi-bin/viewcvs.cgi/*checkout*/Hybrid/Testing/Tcl/TestICPTransform.tcl?root=VTK&content-type=text/plain>for
an example of using the filter.

Note that if you are trying to register a patch to a whole, then you may
need to manually move your patch close to the correct position and call
StartByMatchingCentroidsOff<http://www.vtk.org/doc/nightly/html/classvtkIterativeClosestPointTransform.html#295d56e80cf959f5cde91201371b7afd>()
on the ICP transform before starting the registration.  Otherwise the
registration is likely to get stuck in a local minima.

- Wes


On 5/4/07, M.Salman at surrey.ac.uk <M.Salman at surrey.ac.uk> wrote:
>
>  Thanks Turner,
> I understand what you are saying but I don't know how to code this, Can
> you please elaborate it a bit more ? I really appreciate your help.
>
> Best Regards,
>
> M.Salman
> University of Surrey,U.K.
>
>
>
>
>
> -----Original Message-----
> From: Wes Turner [mailto:wes.turner at kitware.com <wes.turner at kitware.com>]
> Sent: Thu 03/05/2007 4:09 PM
> To: Salman M Mr (PG/R - Electronic Eng)
> Cc: vtkusers at vtk.org
> Subject: Re: [vtkusers] can we match a cropped version of a model againt
> the original using the vtkIterativeClosestPointTransform?
>
> The source and target landmarks need to have the same number of points for
> the point-to-point metric to work.  However, the number of points in the
> two
> models you are registering need not (and generally won't) be the same.
> The
> trick is to subsample from the source and to pick the closest points on
> the
> target as the matches.  For matching patches to their appropriate
> location,
> you may have to find a starting position sufficiently close to the correct
> position of the patch to ensure that you do not get trapped in a local
> minima.
>
> - Wes
>
> On 5/3/07, M.Salman at surrey.ac.uk <M.Salman at surrey.ac.uk> wrote:
> >
> >
> > Hi all,
> > The cropped version of a model doesnt necessarily consists of equal
> number
> > of points as of the original.While vtkIterativeClosestPointTransform
> makes
> > use of  vtkLandmarkTransform internally to compute the best fit which
> > further calls for equal number of points in both landmarks.
> > "Call SetSourceLandmarks and SetTargetLandmarks to specify the two sets
> of
> > landmarks, ensure they have the same number of points"
> >
> > Can we still use the vtkIterativeClosestPointTransform to match the
> > cropped version to the original model...
> > Any thoughts?
> >
> > Regards,
> >
> > M.Salman
> > University of Surrey,U.K.
> >
> >
> >
> >
> > _______________________________________________
> > This is the private VTK discussion list.
> > Please keep messages on-topic. Check the FAQ at:
> > http://www.vtk.org/Wiki/VTK_FAQ
> > Follow this link to subscribe/unsubscribe:
> > http://www.vtk.org/mailman/listinfo/vtkusers
> >
> >
>
>
> --
> Wesley D. Turner, Ph.D.
> Kitware, Inc.
> R&D Engineer
> 28 Corporate Drive
> Clifton Park, NY 12065-8662
> Phone: 518-371-3971 x120
>
>


-- 
Wesley D. Turner, Ph.D.
Kitware, Inc.
R&D Engineer
28 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-371-3971 x120
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.vtk.org/pipermail/vtkusers/attachments/20070504/efeb911e/attachment.htm>