When I have a plaster cast of a mandibula and rest it upside down on a table, in most cases it will be in a stable position.

In real life, I do not have a plaster cast, but a 3D scan, i.e. a surface mesh. I want to transform the meshes to the position they would be in if they were resting on a table.

Why: we have 3D scans of mandibulae, but some of them are slightly misaligned against the axis because the patient did not sit straight. For easier comparison, we would like to make them look more similar. And in case someone suggests "registration": we do that anyway.

Can someone point me to an algorithm? Any keyword I can use to google?

  • 2
    $\begingroup$ The only thing I know of that sounds remotely related to your question is this paper from SIGGRAPH 2013: igl.ethz.ch/projects/make-it-stand/… $\endgroup$ – IneQuation Dec 3 '16 at 12:49
  • $\begingroup$ Sorry, I know, could not get the title into one word. However, I believe both are the same problem. $\endgroup$ – Dieter Menne Dec 3 '16 at 16:13
  • $\begingroup$ I've edited to hopefully make the title and body match. Please revert anything which changes your intention. $\endgroup$ – trichoplax Dec 4 '16 at 0:02
  • $\begingroup$ By using a principal axis rotation, and some shifts, I got what I wanted. Only works because of the special form of a mandibula-scan. $\endgroup$ – Dieter Menne Dec 4 '16 at 11:08
  • $\begingroup$ If you solved your problem, feel free to answer your own question. $\endgroup$ – Christian Rau Oct 1 '17 at 20:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.