Fully Dynamic Delaunay Triangulation in Logarithmic Expected Time per Operation (1992)(Make Corrections)(19 citations) Olivier Devillers, Stefan Meiser, Monique Teillaud
Abstract: The Delaunay Tree is a hierarchical data structure that has been introduced
in [7] and analyzed in [6, 4]. For a given set of sites S in the plane and an order
of insertion for these sites, the Delaunay Tree stores all the successive Delaunay
triangulations. As proved before, the Delaunay Tree allows the insertion of a site in
logarithmic expected time and linear expected space, when the insertion sequence
is randomized.
In this paper, we describe an algorithm maintaining the Delaunay... (Update)
.... of insertion, of ensuring an O(log 2 n) location time for any point, and of allowing deletions in an easier way than the Delaunay tree [DMT92]. However, the additional memory is still important and the location structure is not especially simple. In 1996, Mcke, Saias and Zhu...
...with 6 unknowns, and a unique solution if the points are noncollinear. Thus, we make use of the Delaunay triangulation as implemented by [3] on the first frame and triangulate the second frame by points corresponding. Furthermore we assume that (C u ; C v ; a; b; c; d) are...
O. Devillers, S. Meiser, and M. Teillaud. Fully dynamic Delaunay triangulation in logarithmic expected time per operation. In Proc. 2nd Workshop Algorithms Data Struct., volume 519 of Lecture Notes in Computer Science, pages 42--53. Springer-Verlag, 1991. http://citeseer.nj.nec.com/article/devillers92fully.html More
@techreport{ devillers90fully,
author = "O. Devillers and S. Meiser and M. Teillaud",
title = "Fully dynamic Delaunay triangulation in logarithmic expected time per operation",
number = "RR-1349",
year = "1990",
url = "citeseer.nj.nec.com/article/devillers92fully.html" }