Dynamic remeshing script with Plankton and Kangaroo

Replies to This Discussion

Permalink Reply by jjsolly on January 30, 2014 at 8:22am

That's nice. 

Based on the code shown to date i guess you'd feed the crv into the c# component so you could measure to it... i hadn't thought of that. 

In fact - based on that approach you could simply use the curve instead of the naked-edges of the mesh for then finding curvature. i.e. for every point considered, find the closest point on the curve and evaluate both the distance away and the curvature there. Then use those as the edge-length weighting factors. 

Permalink Reply by Daniel Piker on January 30, 2014 at 8:34am

Good question Will - I guess at the boundary itself one could split/collapse based on this chord-midpt-to-curve distance, but then there would still need to be something that blended these edge lengths smoothly towards the interior of the mesh.

Perhaps each edge could just look at the average of the lengths of connected edges and take that as a target length (possibly adding some small factor so that it got bigger where the detail wasn't needed).

Permalink Reply by Christian Schmidts on April 28, 2014 at 4:22am

Hello Daniel,

I just tried to achieve something similar to this (proximity to the boundary as a sizing function) by inputting points along the boundaries with small values for sizing and points on the interior with larger values.

It works more or less but the mesh is quite unstable and also the transition doesn't look as smooth as yours.

Therefore i wanted to ask which technique did you use and if should rather tweak the input mesh or the parameters of the script to improve it?

Edit: Sorry i just saw your explanation below after the post. That explains the function that you use. So I guess this is a good example where vertex colors as a sizing function would work great..

Permalink Reply by Nik Willmore on June 8, 2014 at 6:33pm

But how, exactly? I found the math for a related 2D version called Chew's Second Algorithm, and a C# port of it, so now I guess I have to learn how to make my own Grasshopper component. Why are simple elegant meshes not built into Rhino and Grasshopper already?!

https://triangle.codeplex.com

Permalink Reply by David Stasiuk on January 28, 2014 at 6:35am

nice with the naked edge valence adjuster!

Permalink Reply by cdelab on December 3, 2015 at 4:48am

Hi Daniel or other guru, 

I'm running the examples for remeshing and always I got the same error. The index was out of range: the line is: if((Visited[P.Halfedges[2 * i].StartVertex] == false) 

How I could resolve the issue?

thanks in advance

Carlos.

Permalink Reply by Contestables on January 20, 2020 at 11:00am

Hi Daniel,

maybe silly question: there is a way to change target valency on some edges of a mesh through some new functionalities of meshmachine or with another tool?

Thank you,

Marinella

Permalink Reply by Wieland Schmidt on January 30, 2014 at 1:04am

Daniel,

thank you for letting us play with the dynamic remeshing script, it works great! I have been looking forward to this since a long time. Is it possible to test cytoskeleton as well?

Wieland

Permalink Reply by Daniel Piker on January 30, 2014 at 3:29am

Hi Wieland,

Thanks for reminding me - I'd forgotten I still didn't actually post the script - will do so now.

Permalink Reply by Robert Harvey on March 15, 2014 at 9:10am

This is great, thank you for your hard work Daniel.  

Im interested in using this in conjunction with Hoopsnake, is it therefore possible to perform a predetermined number of iterations as opposed to using the timer component?  Im pulling mesh vertices incrementally to other geometry and remeshing at each stage, it would be great if I could streamline this process a bit using Hoopsnake.

Thanks

Permalink Reply by Will Pearson on March 15, 2014 at 9:19am

Hi Robert. Are you interested in the results at each step, or just the final result after a predetermined number of iterations?

Permalink Reply by Robert Harvey on March 15, 2014 at 9:24am

Thanks for replying so soon.

I'm not interested in each iteration within the remeshing script, the results produced are good enough very quickly.  Previously I've used Kangaroo without the timer and was hoping to do the same here and to be able to experiment a bit with the number of iterations performed.

• ‹ Previous
• 1
• 2
• 3
• 4
• 5
• 6
• …
• 8
• Next ›
• Page