Comments on: Poisson Disk Sampling in Flash http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/ Experiences in AS3 Fri, 09 Jul 2010 04:49:54 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Kim Christiansen http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2461 Kim Christiansen Thu, 10 Sep 2009 18:07:42 +0000 http://www.bigsource.de/?p=129#comment-2461 Mario, this sounds very interesting! Is your image foam algorithm open source? Another nice effect of the poisson disk method of finding a new point is the way it looks when you decrease the number of points: they are placed in groups, like f.x. mold. Mario, this sounds very interesting! Is your image foam algorithm open source?
Another nice effect of the poisson disk method of finding a new point is the way it looks when you decrease the number of points: they are placed in groups, like f.x. mold.

]]> By: Mario Klingemann http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2460 Mario Klingemann Thu, 10 Sep 2009 16:52:56 +0000 http://www.bigsource.de/?p=129#comment-2460 Yes it's a bit different but I also use several control maps that define the size, rotation and other properties of an object at a certain position of the image. So in theory I could limit it in a way that there is no growing at all but a fixed size and only an "either it fits or it gets thrown away" condition. In that case it would take a bit longer to fill the whole image and some areas might not get filled at all. Your method of finding a new point sounds very good. Since with this space filling algos the biggest problem is to find empty spots once the image has becomes crowded. Yes it’s a bit different but I also use several control maps that define the size, rotation and other properties of an object at a certain position of the image. So in theory I could limit it in a way that there is no growing at all but a fixed size and only an “either it fits or it gets thrown away” condition. In that case it would take a bit longer to fill the whole image and some areas might not get filled at all.

Your method of finding a new point sounds very good. Since with this space filling algos the biggest problem is to find empty spots once the image has becomes crowded.

]]> By: Kim Christiansen http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2459 Kim Christiansen Thu, 10 Sep 2009 12:27:06 +0000 http://www.bigsource.de/?p=129#comment-2459 I really like the idea of adding and hittesting shapes instead of circles. And the results are awesome! But i think the image foam algorithm is a bit different. The image foam lets the shapes "grow" until it hits an shape, right? In the poisson disk algorithm the shape already have an size on a given coordinate (defined out of an image). If the shape hits an other shape, it will not be added to the list and a new random point will be selected. The coordinates of the new point is calculated out of the position of the previous point and its radius. The new point will be near the previous point. I really like the idea of adding and hittesting shapes instead of circles. And the results are awesome!
But i think the image foam algorithm is a bit different. The image foam lets the shapes “grow” until it hits an shape, right? In the poisson disk algorithm the shape already have an size on a given coordinate (defined out of an image). If the shape hits an other shape, it will not be added to the list and a new random point will be selected. The coordinates of the new point is calculated out of the position of the previous point and its radius. The new point will be near the previous point.

]]> By: Mario Klingemann http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2458 Mario Klingemann Thu, 10 Sep 2009 10:54:18 +0000 http://www.bigsource.de/?p=129#comment-2458 That looks like the same technique I am using in my Image Foam pieces: http://www.flickr.com/photos/quasimondo/sets/72157600012142566/ That looks like the same technique I am using in my Image Foam pieces: http://www.flickr.com/photos/quasimondo/sets/72157600012142566/

]]> By: nicoptere http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2457 nicoptere Fri, 04 Sep 2009 15:55:01 +0000 http://www.bigsource.de/?p=129#comment-2457 might prove really useful. thanks for sharing :) might prove really useful.
thanks for sharing :)

]]> By: localToGlobal » news review -> 36th week of 2009 http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2455 localToGlobal » news review -> 36th week of 2009 Fri, 04 Sep 2009 15:38:52 +0000 http://www.bigsource.de/?p=129#comment-2455 [...] Poisson Disk Sampling in Flash by Kim Christiansen [...] [...] Poisson Disk Sampling in Flash by Kim Christiansen [...]

]]> By: Tizzle http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2454 Tizzle Fri, 04 Sep 2009 11:14:31 +0000 http://www.bigsource.de/?p=129#comment-2454 I likes. I did a painterly render algorithm in Processing a while ago and thought about integrating Poisson Disk Sampling as a possibility to place the brush strokes. I decided it's too much effort so i ended up with a randomized jittered Grid which does the job quite well. I'll take a look at your sources, maybe i can steal your Sampling ;) I likes. I did a painterly render algorithm in Processing a while ago and thought about integrating Poisson Disk Sampling as a possibility to place the brush strokes. I decided it’s too much effort so i ended up with a randomized jittered Grid which does the job quite well. I’ll take a look at your sources, maybe i can steal your Sampling ;)

]]> By: Jens Franke http://www.bigsource.de/index.php/2009/09/04/poisson-disk-sampling-in-flash/comment-page-1/#comment-2453 Jens Franke Fri, 04 Sep 2009 11:05:42 +0000 http://www.bigsource.de/?p=129#comment-2453 Nice one! I really like the idea. Cheers Jens Nice one! I really like the idea.

Cheers
Jens

]]>