<blockquote style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;" class="gmail_quote"><div class="gmail_quote">Hi David,<br>
<br>
I'm using VTK 5.4.2 and compiled it on my 64-bit Gnu/Linux machine running kernel 2.6.25.18.<br>
<br>
See attached the files needed.<br>
One should be able to run the algorithm through the run_VTKICP.py file - it basically constructs a class that is in the other file and runs the ICP algo.<div class="im"><br>
<br>
ciao,<br>
Sean.<br></div></div></blockquote><br>Sean,<br><br>I also get the eigenfunction errors when running your python conversion of the ICP example. I looked carefully at your code. The only thing I saw that was different was the second target point, but even when I changed it to:<br>
id = targetPoints.InsertNextPoint(0.0, 1.0, 0.0)<br><br>to match my c++ example, it still produces the eigenfunction error. This could be a bug in the python wrapper? Can one of the experts take a look?<br><br>As I said before, this error is produced *correctly* (although it could probably be caught and reported in a much easier to understand way - like "degenerate transformation") when the transformation is not possible, but since we know that it IS possible in this case, I'd guess it is a problem with how python is passing along the point or polydata arrays? I have no idea how to check that though.<br>
<br clear="all">Thanks,<br><br>David<br>
<br>