Papers for

graph visualization developers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Arrangements of lines and pseudolines with maximum triangles exceed known bounds

Non-simple Arrangements of Lines and Pseudolines with Many Triangles

Abstract: We study affine arrangements of an even number $n$ of lines and pseudolines with many triangles. For simple ones, Blanc's bound limits the number of triangular faces to $\lfloor n(n-5/2)/3\rfloor$. Adapting the Bartholdi-Blanc-Loisel doubling to the even case, we construct an infinite series of arrangements of $n=6\cdot2^t+2$ ($8,14,26,50,\ldots$) straight lines with two triple points and $(n(n-5/2)+1)/3$ triangles, one more than the simple bound allows. We obtain exhaustive enumerations of triangle-maximal pseudoline arrangements, simple or not, for even $n\le12$. For $n=6,8,10,12,14$ the maximum number of triangles, for both lines and pseudolines, is $7,15,25,38,54$; for $n=8,12,14$ every optimal arrangement has multiple points. All optimal arrangements for even $n\le12$ are realizable by straight lines, except two arrangements of 10 pseudolines, which are not stretchable. For the odd value $n=11$, no arrangement with $33$ triangles exists even with multiple points. We also obtain even arrangements by adding a line to odd ones. For $n\equiv2\pmod 6$ this again gives one triangle more than the simple bound, in pseudolines up to $n=44$ and in straight lines for $n=20$ and $n=32$. Doubling these pseudoline arrangements gives members of the series $18\cdot2^t+2$ and $30\cdot2^t+2$ with the same excess, up to $n=146$ and $n=242$, respectively; those with $n=38$ and $n=74$ are realizable by straight lines. These results lead us to conjecture that an arrangement of an even number $n$ of pseudolines has at most $\lfloor n(n-2)/3\rfloor-\lfloor n/6\rfloor$ triangles.

Mon 2 NovComputational Geometry
The gist
Arrangements of lines or similar curves can form many triangular regions where they intersect. Previous limits suggested how many triangles are possible when intersections are simple, but the authors found examples with special multiple intersections that create even more triangles. They identified precise numbers and patterns for these arrangements, including some that cannot be made with only straight lines. Their work leads to a new guess about the absolute maximum number of triangles possible in such arrangements.
Open → 2611.03096v1

Graph automorphisms matched with drawing symmetries in plane

Realizing graph automorphisms as geometric symmetries of straight line drawings

Abstract: Given a finite graph $G$, we characterize the automorphisms and automorphism groups of $G$ that can be realized as geometric symmetries and symmetry groups of straight line drawings of $G$ in the plane. In doing so, we correct a previous version of this characterization from the literature. As an application, we show that the graphs recently constructed by the first two authors to disprove the Polycirculant Conjecture, which have orders $16{,}464^k$ for each positive integer $k$, are the first known vertex-transitive graphs with no non-trivial realizable geometric symmetries. We also construct infinite families of graphs that exhibit all possible kinds of geometric symmetry.

Mon 26 OctComputational Geometry
The gist
Graphs are made of dots and lines, and sometimes the graph can be matched with itself in interesting ways called automorphisms. This paper figures out which of these self-matchings can also appear as symmetrical drawings made with straight lines on a flat surface. They fixed errors from earlier work and show examples of graphs that cannot have any non-trivial straight-line symmetries. The authors also built families of graphs that show all kinds of possible symmetrical patterns in their drawings.
Open → 2610.36394v1