What is the best way to do this?

Hey guys, new user here

I’m trying to figure out the best way to do a high relief model of this image. It’s a very complicated thing for me, because I’ve been learning how to use Rhino just for 2 weeks. I already know a lot about solids and surfaces, but I can’t even start modeling this… My objective is modeling this image over a medal, for 3D printing and laser marking. It needs to be a high relief. That’s quite difficult.

I really need your help, guys! Please! Does anyone know how can I do this?

I think GH is the best way to do it. For the most part you take an [Image Sampler] component, a mesh with a lot of vertices, and then the image sampler’s output is used to control the z values of the mesh’s vertices. Here is a thread with links and examples.

Sampling 32 million pixels is going to take some time though…

Hi Hayashi - do you want to model it as some kind relief or as a fully 3d object?

-Pascal

You could open it in Adobe illustrator and trace the pattern to simplify it a little in order to give you cleaner lines before you make a start.

If it’s a case of having sharp edges it won’t take very long. You can import the PDF and use the outlines to extrude the shape as a solid.

If it has to have a rounded look and feel like worn bas relief then it maybe better to do it with grasshopper as suggested or a displacement map: https://youtu.be/r_CB6-OjCLU

Andy

Hi pascal

Actually I want to do something like this:

Basically, transform a 2D sample in a relief 3D model.

Hi Marcus, thank you for your suggestion. I’m going to take a look to Grasshooper.

Thank you, 2DCube. I’ll try it.

Hi Hayashi - it is not an ideal image to work from - you might want to blur it a little, but Heightfield with the Mesh option may be a start.

-Pascal

Ok… This what I got using Heightfield:

I know I need a better image… But when I get it, what exactly should I do with Heightfield and Mesh?

Hi Hayashi - I think I’d output a mesh and not a surface and a much lower height in the dialog, but also the image itself is only a 2 color bitmap - it should be a grayscale with some blurring at the edges between black and white.

Grayscale, blurred a little, reduced in size, mesh resolution set to match the pixels in the image (i.e. one vertex per pixel)

@Hayashi - here’s my quickly edited version of your image-

The image is grayscale, blurred to soften the transition between black and white (i.e. round the edges) and 1250 by 1506 pixels (1/4 the size of the original). Heightfiled is set to 1249 by 1596, hoping to land one mesh vertex on each pixel of the image.

-Pascal

2 Likes

omg how did you do that?

So I need to transform this image in a grayscale pictureframe in order to get something like this?

Thank you so much!!