Mean shift for graph bundling

Ozan Ersoy, Christophe Hurter, Alexandru Telea
Institute Johann Bernoulli, University of Groningen, the Netherlands


   title={Mean shift for graph bundling},

   author={Ersoy, Ozan and Hurter, Christophe and Telea, Alexandru},



Download Download (PDF)   View View   Source Source   



We present a fast and simple adaption of the well-known mean shift technique for image segmentation to compute bundled layouts of general graphs. For this, we first transform a given graph drawing into a density map using kernel density estimation. Next, we apply the equivalent of mean shift segmentation on this image, i.e. sharpen the image my moving the drawn edges upstream in the density’s gradient. We implement our method using standard graphics acceleration techniques. Our results are similar to state-of-the-art graph bundling methods but require a fraction of their cost. We demonstrate our method on several large graphs.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2021 hgpu.org

All rights belong to the respective authors

Contact us: