Page 1 of 1

Subdivision, normals, and signed distance explanations

Posted: Mon Jul 20, 2020 8:32 pm
by Sarah m Al
Hello,

I am working on a research project that involves assessing the distance between 2 corresponding meshes using CloudCompare, while working with an experienced colleague, I was advised to always invert the normals for better viewing of the mesh, subdivide the files, and maintain the signed values.

That certainly helped me proceed with my project and my distance readings were reasonable, however, for publication purposes, I have the following questions:
1- If my object has sharp and rounded edges, does subdividing the mesh helps with the alignment or the file resolution?
2- Does inverting the normals only help with color visualization, or does it also help with the alignment of both meshes?
3- When my distance readings are showing signed values around the same areas, with the 2 meshes colliding into each other when I know in real life that both objects do no interlock. Is it possible to explain the signed values as a direction only?
For example, 2 areas are showing -0.05 and +0.05 are they both considered a distance of 0.05? or does the negative value means the distance is smaller and that's why the meshes are interlocking? how can I interpret the signed distances in my paper?

Thank you for your help.

Re: Subdivision, normals, and signed distance explanations

Posted: Tue Jul 21, 2020 3:20 pm
by daniel
1. If the subdivision doesn't change the shape, then it won't change anything (it will just slow the process down)

2. Nope, only display indeed

3. Yes, negative distances are indeed an indication the the point (vertex here) is 'behind/below' the nearest triangle. Which generally means that the meshes are interlocking indeed.

Re: Subdivision, normals, and signed distance explanations

Posted: Sat Feb 19, 2022 9:34 am
by Sarah m Al
Thank you Daniel for your explained replies.

I am at the stage of submitting my paper to a scientific journal, I have referenced the CloudCompare user manual as "Daniel Girardeau-Montaut, AB and RM, CloudCompare user manual, 2020, GPL Software, v2.6.1."

However, the publishers requested the full names of the author's AB and RM and I was unable to find that. Would you mind clarifying the full author names in order for me to proceed with the citation and submission?

Thank you

Re: Subdivision, normals, and signed distance explanations

Posted: Sat Feb 19, 2022 10:09 am
by daniel
Hi, the normal citation format is simpler:
https://www.cloudcompare.org/forum/view ... ?f=2&t=146

And the real user manual is the wiki: https://www.cloudcompare.org/doc/wiki/i ... =Main_Page