generative modeling for Rhino
I started using Grasshopper 5 weeks ago at uni and I am trying to figure out how to recreate the image below. My initial reaction was the use the image sampler and output it to a circle component however I need to associate certain circles with the icons that you see below. Any help would be greatly appreciated!
I don't know about this project but from what I can see in your picture your "icons" seems to be "uniform size white circles at >0.5 brightness" with some "letters the size of the brightness 0 to 1" on top of it (size/colour inverted at white circles). Does that answer your question?
IIT by OMA! I'd like to know the answer to this question, too. And branching from it, how do you map a picture using circles of different sizes?
Picture using circles of different sizes: please see this video: http://www.youtube.com/watch?v=ym_jI0BSD6M
To summerize, plug the image sampler output into the radius of the circles. (Darker = smaller circles, brighter = bigger circles!)
Thank you for the helpful link. Allow me to challenge you a little bit more: I generated the image with circles, and move them to vertical panels that can happen anywhere in the model.
How do I make the panels perforated? Do I need to make an algorithm that recognize the boundary of each panel, find all circles inside the boundary, then generate a planar surface using the panel edges and the circles inside? Is there an easier way?
Picture from http://www.flickr.com/photos/hellebelle/2417598192/
As far as I can tell in OMA's example the white circles are always the same size but appear only at brightness >0.5. It is then filled/modified with various symbols or icons with different apperant "brightness" (or darkness when inside a circle) determined by their surface area.
So for example a white pixel with brightness =1 will be a white circle with nothing in it. A brightness 0.9 pixel will be a white circle with a small icon in it (thus taking away a little whiteness of the circle). A brightness 0.6 pixel will be a white circle with a huge icon in it taking away the majority of the circle's area. A 0.4 will be a large icon in white without circle. A 0.1 will be a tiny white icon. A zero brightness pixel will be totally transparent (nothing).
So in order to replicate OMA's creation you just need to prepare a collection of icons with different sizes and place them into the grid accordingly. I think finding (or drawing) the icons will be the hardest part because once you have got your icons all you need to do is to sort them by area size (or average them individually in photoshop and pick up the grey value and sort them if you know what I mean) and just grasshopper them with the logic I have just described.
Thanks for your response.
Problem is i need to fabricate this so all of the line work needs to be vectors - or so i am told. Is there no way to plug the image sampler into a grid of points that then associates the icons with the darkness or lightness of the source image. If all of the circles ar the same size as you say then pugging an image sampler into a circle will leave me with several different circle sizes...
Also its the "grasshopper" logic that i dont understand.... Only new a this so my knowledge is limited...
Thankyou for your help! Much appreciated!!
use remap values component to make it a "real" negative. (bounds + domain components + remap values)
This discussion could also be helpful background in your quest...
Letter & ASCII image generation: