Wenzheng,
It seems that the procrustes scaling has been broken for several
versions. We will issue a fix shortly. You can either build it from
source when ready or use a new binary distribution when we create one.
I will point out that your sample data has another problem in that the
last image is very tiny, only a few voxels are inside the zero level
of the distance transform. Even with a working procrustes, it is not
able to handle this degenerate shape.
Thanks,
Alan
On Wed, Nov 20, 2019 at 8:51 AM wenzheng tao <wztao@cs.utah.edu>
wrote:
Hi Alan,https://github.com/SCIInstitute/ShapeWorks/releases/download/v5.1/shapeworks-5.1-linux-legacy.zip
Thanks for your suggestions! 'shapeworks-5.2.1' works on another
system
of 'openSUSE Leap', system info attached as 'systeminfo2.out'. I
think
the old system brings the problem with the latest version.
However, the scaling of ShapeWorksRun does not seem to be
functioning,
and this is why I want to try the latest version.
Please find attached a zip file containing the test for scaling. It
contains segmentation, distance transform of ellipsoid samples and
xml.
I have set the 'procrustes_scaling' to 1 and 'procrustes_interval'
to 1.
And I found the surface of the coorespondences are not of the same
size,
while I thought that they should be after applying
procrustes_scaling. I
am reconstructing the surface from correspondences using 'ball
pivot'
method of meshlab.
Could you have a look at the test experiment and let me know how to
do
the scaling with ShapwWorks? Thanks for your time!
Wenzheng
On 2019-11-19 23:17, Alan Morris wrote:
Hi Wenzheng,work
RHEL 7 is quite old (>5 years). You can try out previous release
where we built a legacy linux set of binaries that I believe will
for you.
https://github.com/SCIInstitute/ShapeWorks/releases/download/v5.2.1/shapeworks-5.2.1-linux.zipfiles
Thanks,
Alan
On Tue, Nov 19, 2019 at 3:31 PM Shireen Elhabian
<shireen@sci.utah.edu> wrote:
Hi Cam and Alan,
Would you mind looking into this?
thanks and best regards
Shireen
------------------------------------------
Shireen Elhabian, M.Sc., Ph.D.
Research Assistant Professor
School of Computing
Scientific Computing and Imaging Institute
University of Utah
WEB 2815
72 Central Campus Drive, Salt Lake City, UT, 84112
Phone (801) 587-3206
Fax (801) 585-6513
Home Page: http://www.sci.utah.edu/~shireen
On Nov 19, 2019, at 3:16 PM, wenzheng tao <wztao@cs.utah.edu>wrote:
Hi,
I am using the latest ShapeWorks but got the error that some
wrong?are not found. Could you help me find out what might be going
file
I am using 'Red Hat Enterprise Linux Server', as in attached
'systeminfo.out'.
from:
I downloaded the latest linux release shapeworks-5.2.1-linux.zip
installed'~/shapeworks-5.2.1-linux/bin' into PATH.
Then I unzipped it and add the binary directory
attached file 'ShapeWorksRun1.out'.
I ran 'ShapeWorksRun xxx.xml' and got the output with errors in
I also activated the shapeworks conda environment that I
gotwith an older version before.I ran 'ShapeWorksRun xxx.xml' and
releasethe output with the same errors in attached file
'ShapeWorksRun2.out'.
Besides, I am able to run python example scripts from old
and old binary 'ShapeWorksRun5.0'.
--
Best Regards!
===============================================
Wenzheng Tao
School of Computing
University of Utah
===============================================
<systeminfo.out><ShapeWorksRun2.out><ShapeWorksRun1.out>
Archive powered by MHonArc 2.6.18.