Force-directed Lombardi-style graph drawing

Roman Chernobelskiy, Kathryn I. Cunningham, Michael T. Goodrich, Stephen G. Kobourov, Lowell Trott

Research output: Chapter in Book/Report/Conference proceedingConference contribution

35 Scopus citations

Abstract

A Lombardi drawing of a graph is one in which vertices are represented as points, edges are represented as circular arcs between their endpoints, and every vertex has perfect angular resolution (equal angles between consecutive edges, as measured by the tangents to the circular arcs at the vertex). We describe two algorithms that create "Lombardi-style" drawings (which we also call near-Lombardi drawings), in which all edges are still circular arcs, but some vertices may not have perfect angular resolution. Both of these algorithms take a force-directed, spring-embedding approach, with one using forces at edge tangents to produce curved edges and the other using dummy vertices on edges for this purpose. As we show, these approaches produce near-Lombardi drawings, with one being slightly better at achieving near-perfect angular resolution and the other being slightly better at balancing edge placements.

Original languageEnglish (US)
Title of host publicationGraph Drawing - 19th International Symposium, GD 2011, Revised Selected Papers
Pages320-331
Number of pages12
DOIs
StatePublished - 2012
Event19th International Symposium on Graph Drawing, GD 2011 - Eindhoven, Netherlands
Duration: Sep 21 2011Sep 23 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7034 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other19th International Symposium on Graph Drawing, GD 2011
Country/TerritoryNetherlands
CityEindhoven
Period9/21/119/23/11

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Force-directed Lombardi-style graph drawing'. Together they form a unique fingerprint.

Cite this