Re: Force-Directed Edge Bundling
Posted: 21 Mar 2012 21:24
Hi Taras,
I am glad you are applying to FDEB proposal.
You have previously completed several features for Gephi and have good algorithms background (I am sure you do better than me
)
I need to get a better understanding of the barnes-hut optimization (which works really good for force atlas 2), but I think they refer to using it for quickly discarding edges based on position compatibility since if it is close to 0, the other compatibility measures would not matter.
Anyway I think we would start to implement basic FDEB, when it works create some visualization and when things are good start to think about optimizations and tuning up the algorithm.
About gradients, that's is something I was worried too. For SVG (and pdf I guess) rendering there seems to be support http://www.w3schools.com/svg/svg_grad_linear.asp but for processing rendering I think it has to be done manually http://wiki.processing.org/w/Vertical_Gradient
Eduardo
I am glad you are applying to FDEB proposal.
You have previously completed several features for Gephi and have good algorithms background (I am sure you do better than me
)I need to get a better understanding of the barnes-hut optimization (which works really good for force atlas 2), but I think they refer to using it for quickly discarding edges based on position compatibility since if it is close to 0, the other compatibility measures would not matter.
Anyway I think we would start to implement basic FDEB, when it works create some visualization and when things are good start to think about optimizations and tuning up the algorithm.
About gradients, that's is something I was worried too. For SVG (and pdf I guess) rendering there seems to be support http://www.w3schools.com/svg/svg_grad_linear.asp but for processing rendering I think it has to be done manually http://wiki.processing.org/w/Vertical_Gradient
Eduardo