shapeworks-users

Text archives Help


Re: [shapeworks-users] ShapeWorks


Chronological Thread 
  • From: manasi <manasi.datar@gmail.com>
  • To: Sangwoo Leem <sleem@uwo.ca>
  • Cc: shapeworks-users@sci.utah.edu
  • Subject: Re: [shapeworks-users] ShapeWorks
  • Date: Fri, 13 Jun 2014 09:22:15 +0200
Hi Sangwoo

It is great that you were able to use ShapeWorks for your research. I hope it will be a useful tool for you.

An incorrect mean shape could result from two problems: incorrect correspondences or reconstruction issues.
I presume you are able to use the ShapeWorksView tool to look at individual shapes reconstructed from the point correspondences. Would you please check if the correspondences match across shapes? An easy, visual test would be to flip through individual shapes and see if similarly colored particles and their neighborhoods stay constant across shapes.

If you have good correspondences, the problem lies with the reconstruction of the mean shape. ShapeWorks natively uses the reconstruction filter from VTK and sometimes requires tuning of the two parameters (neighborhood size and smoothing) to get a good reconstruction of the mean shape. In the past, I have used other reconstruction methods (e.g. powercrust) to get better-looking mean shapes. Perhaps you could look that up...

HTH
Regards
manasi

----------
विज्ञान प्रचिति, गुरू प्रचिति, आत्म प्रचिति
translation: understand the science, learn from the teacher, assimilate for yourself :)


On Fri, Jun 13, 2014 at 2:22 AM, Sangwoo Leem <sleem@uwo.ca> wrote:
Dear Dr. Datar,

My name is Sangwoo Leem and I am currently a student at the University of Western Ontario, Canada undertaking research at the Robarts Institute. I am emailing you after having read your paper on "Toward an understanding of the short bone phenotype associated with multiple osteochondromas" and seeing that you were successfully able to use the ShapeWorks software program. 

I am working on a project analyzing mice skull shapes and my project entails that I work with the ShapeWorks software program that was developed at the University of Utah. 
I was able to get a majority of the software program running well but I am currently having trouble generating a well-defined Mean structure of the skull images that I input, despite each individual skull being generated nicely. The mean structure looks jagged/spiky and nothing like what the mean structure SHOULD look like! 

I was wondering if you would be so kind as to assist me in the problem. I am hoping to generate a smooth, well-defined mean shape from which I can continue my research. 

Thanks for your time and I hope to hear from you soon.

Best Regards,
Sangwoo Leem


  • Re: [shapeworks-users] ShapeWorks, manasi, 06/13/2014

Archive powered by MHonArc 2.6.18.

Top of page