Geometric Separators and the Parabolic Lift
|
|
- Judith Todd
- 5 years ago
- Views:
Transcription
1 CCCG 13, Waterloo, Ontario, August 8 1, 13 Geometric Separators and the Parabolic Lit Donald R. Sheehy Abstract A geometric separator or a set U o n geometric objects (usually balls) is a small (sublinear in n) subset whose removal disconnects the intersection graph o U into roughly equal sized parts. These separators provide a natural way to do divide and conquer in geometric settings. A particularly nice geometric separator algorithm originally introduced by Miller and Thurston has three steps: compute a centerpoint in a space o one dimension higher than the input, compute a conormal transormation that centers the centerpoint, and inally, use the computed transormation to sample a sphere in the original space. The output separator is the subset o S intersecting this sphere. It is both simple and elegant. We show that a change o perspective (literally) can make this algorithm even simpler by eliminating the entire middle step. By computing the centerpoint o the points lited onto a paraboloid rather than using the stereographic map as in the original method, one can sample the desired sphere directly, without computing the conormal transormation. 1 Geometric Separators n d+ A spherical geometric separator o a collection o n balls in R d n is a sphere S that has at least d+ balls centered inside, at least centered outside, and intersects at most O(n 1 1 d ) o them (see Section or a ormal deinition). The existence o such separators in two and three dimensions was established by Miller and Thurston, though their method was quickly adapted to higher dimensions. Across a series o papers, Miller, Thurston, Teng, and Vavasis laid out the theory o geometric separators and their applications to scientiic computing [14, 15, 13, 1, 1. This line o work is a treasure trove or computational geometers as it hinges on a novel trick that combines projective and combinatorial geometry to solve an important algorithmic problem, solving linear systems arising in inite element analysis. More generally, geometric separators give a natural way to do divide and conquer or geometric problems. They have been applied to various nearest neighbor search problems [11 as well as to mesh compression [1. Other variations o geometric separators have been used or the Traveling Salesman and Minimum Steiner Tree problems in geometric settings [18, or or packing and piercing problems [. The Miller-Thurston algorithm or computing a geometric separator maps the n points (the centers o the balls) to a unit d-sphere in R d+1 via a stereographic map. It then computes a centerpoint, which is a geometric generalization o a median (a ormal deinition is given in Section ). There exists a conormal transormation o the points in R d+1 so that this centerpoint will lie exactly at the origin. To output a separator, one samples a random unit vector in R d+1. The hyperplane through the origin normal to this vector intersects the unit d-sphere at a (d 1)-sphere. The output is just the stereographic projection o this (d 1)-sphere back to R d. With high probability, such a sphere will be a geometric separator. The one aspect o this algorithm that was let to the reader, was the linear algebra required to compute the desired conormal transormation. In the original paper, it was simply asserted that it exists. Later papers explained that it can be computed via Householder transormations and cited a textbook on matrix computations. In this paper, we show that this phase o the algorithm is entirely unnecessary. By working initially with the parabolic liting p [ p p, rather than the stereographic map, the desired sphere can be sampled directly. Related work In addition to the previously mentioned work on sphere separators and their applications by various combinations o Miller, Thurston, Teng, and Vavasis, the generalization to other types o contact graphs and other shapes o separators (in particular hypercubes) was perormed by Smith and Wormald [18. This method was reined slightly by Chan to apply separators to geometric hitting set problems [. Eppstein et al. gave a linear-time, deterministic algorithm or inding geometric separators based on core sets [5. Har-Peled gave a simple proo o the existence o geometric separators or interior-disjoint disks in the plane (that easily extends to higher dimensions) [6. Department o Computer Science and Engineering, University o Connecticut, don.r.sheehy@gmail.com
2 5 th Canadian Conerence on Computational Geometry, 13 Deinitions Points We treat points in Euclidean space as column vectors. As the algebra will be in both R d as well as R d+1, we adopt the convention that a boldace vector p is a vector o R d and an overline such as in p indicates a vector in R d+1. In particular, we write to indicate the zero vector in R d. Scalars are italic and when it is useul, we add a subscript d + 1 as in p d+1 to indicate a scalar that is the (d + 1)st coordinate o a vector in R d+1. So, or example, we will write p = [ p p d+1 to indicate that p R d+1 with its irst d coordinates matching those o p and last coordinate equal to p d+1. Whenever we speak o R d as a subspace o R d+1, it is always assumed that we mean the hyperplane {p p d+1 = } in R d+1. Projections Given a point = [ d+1, we deine Π 1 to be the stereographic map rom R d to the sphere centered at with radius d+1. That is, Π 1 (p) is the intersection o the sphere with the line through [ p and the north pole, [ d+1 (other than the pole itsel). Similarly, we deine Π to be the parabolic liting map rom R d to R d+1 that lits p to p = [ p p d+1 so that p lies on the paraboloid with ocal point and directrix {p d+1 = d+1 }. The reason or the notation comes rom the act that the parabola is the limiting case o an ellipse ormed by moving one ocal point to ininity. We will consider more general stereographic projections in Section 4. In all cases, these maps are invertible (except at the north pole). We abuse notation slightly and let Π 1 (S) denote the set {Π 1(p) p S} and similarly or Π. In particular, Π 1 (Rd ) denotes the sphere centered at with radius d+1 and Π (R d ) denotes the paraboloid with ocal point and directrix {p d+1 = d+1 }. Centerpoints Given n points in R d, a centerpoint is a point c R d such that any closed halspace containing more than nd d+1 points also contains c. The existence o centerpoints ollows rom Helly s Theorem and the pigeonhole principle. Let Centerpoint(P ) denote the set o all centerpoints o the set P. It is not known how to ind a centerpoint deterministically in polynomial time or point sets in R d. However, there is an eicient randomized algorithm [3 known or at least twenty years and some recent work on deterministic approximation algorithms [9, 16. Sphere Separators A graph separator is a subset o vertices in a graph whose removal disconnects the graph. Usually, the goal is to ind a small separator that separates the graph into roughly equal sized pieces. The most amous example o graphs admitting small separators is the Planar Separator Theorem o Lipton and Tarjan [8, which states that a separator o size O( n) is always possible or planar graphs that has at least n 3 vertices in each o the resulting components. The early work on separators was directed at solving linear systems by generalized nested dissection, a method or ordering pivots in Cholesky decomposition o sparse, symmetric, positive deinite matrices [7. O particular interest were those linear systems arising in the inite element method. These systems relected the underlying geometric structure o the problem domain and thus it was natural to look to the geometry to ind small separators [1. In particular, it oten suiced to consider various deinitions o intersection graphs o systems o balls. Let B = {B 1,..., B n } be a collection o interior disjoint balls in R d. Let S be a sphere in R d and let B I (S), B E (S), and B O (S) be the subsets o B that are interior to, exterior to, and intersecting S respectively. The ollowing theorem is a combination o the main existence result or geometric separators with the algorithm used to prove existence. We state it this way, to make clear that the geometric challenge or computing a separator this way lies in inding a stereographic projection to a sphere that has a centerpoint at the center o the sphere. Theorem 1 (Sphere Separator Thm.[14, 11, 1) Let B be a collection o n interior disjoint balls with centers P. Let v R d+1 be chosen uniormly rom the unit d-sphere. I is a point in R d+1 such that is a centerpoint o Π 1 (P ) and H = {p v (p ) = } is the hyperplane through normal to v, then the sphere has the property that S = (Π 1 ) 1 (Π 1 (Rd ) H) B O (S) = O(n 1 1/d ), and B I (S), B E (S) with probability at least 1. (d + 1)n d + For ease o exposition, we only give this simplest version o the theorem. However, it has also been proven or k-ply neighborhood systems where the balls are permitted to have up to k wise interior intersections [11 as well as or α-overlap graphs where two balls are considered neighbors i or both balls increasing the radius o one by a actor o α causes them to intersect [1. In these cases, the bounds depend on k and α respectively, but the algorithm is the same and so the results o this paper apply to these versions o the theorem as well. The version stated above, when combined with the Koebe-Andreev-Thurston embedding theorem or planar graphs is strong enough to prove the Planar Separator Theorem.
3 CCCG 13, Waterloo, Ontario, August 8 1, 13 3 The Algorithm When Archimedes quipped that he could move the earth i given a suiciently long lever, he transerred the majority o the work to the lever builders o his day. We will do likewise and assume that we are given a long lever in the orm o an algorithm to eiciently compute a centerpoint o n points in R d+1. Given such an algorithm, the heavy liting will be quite easy, and as with Archimedes, that is entirely the point. Let P = {p 1,..., p n } R d be a set o points. We irst present the Miller-Thurston algorithm or computing a separator and then present the new simpliied version. 3.1 The Miller-Thurston Algorithm Let Π be the stereographic map rom R d to the unit d-sphere centered at the origin in R d+1. First, compute c Centerpoint(Π(P )). Find an orthogonal transormation Q such that Q(c) = [ θ or some θ R. 1 θ Let D = 1+θ I, where I is the identity on Rd. Choose a random unit vector v R d+1 and let S be the d-sphere ormed by intersecting the hyperplane {p v p = } with the unit d-sphere centered at the origin. Output S = Π 1 (Q 1 Π(DΠ 1 (S ))). 3. A Simpler Algorithm Using the Parabolic Lit First, compute c = [ c [ p1 [ pn c d+1 Centerpoint( p 1,..., p n ). Next, choose a random unit vector v = [ v v d+1 R d+1. Let cd+1 c r =. v d+1 Output the sphere S with center (c rv) and radius r. In the improbable case that v d+1 =, the output is just the hyperplane {p v (p c) = }. 3.3 Correctness o the Algorithm The remainder o this section will prove that the algorithm works. According to the results o Miller et al. [1, all we need is a stereographic projection o R d to a d-sphere such that the projected points have a centerpoint at the center o the sphere. The trick presented here is that we will instead show how to ind a parabolic liting that has a centerpoint at the ocal point o the paraboloid. Then, we show that i we have a point such that Π (P ) has a centerpoint at, then Π 1(P ) also has a centerpoint at, thus giving the desired map. For any vector v R d+1 and any p R d, v (Π 1 (p) ) = i v (Π (p) ) =. That is, or sampling spheres, there is no dierence between using the stereographic map or the parabolic liting. In act, we prove a much more general statement about the equivalence o various stereographic projections in Theorem 3. Theorem Let B be a collection o n interior disjoint balls with centers P = {p 1,..., p n } R d and let S be the sphere output by the algorithm in Section 3.. Then, B O (S) = O(n 1 1/d ), and B I (S), B E (S) with probability at least 1. (d + 1)n d + Proo. Using Theorem 1, it will suice to show there exists such that S = (Π 1 ) 1 (Π 1 (Rd ) H) where H = {p v (p ) = } and Centerpoint(Π 1 (P )). The equivalence o dierent stereographic projections proven in Theorem 3 implies that it will suice to prove the same acts replacing the stereographic map Π 1 with the parabolic lit Π. We will show that the ocal point that makes this true is = [ 1 c cd+1 c, where c = [ c c d+1 is the centerpoint o Π (P ) computed in the irst step. First, we show that S = (Π ) 1 (Π (R d ) H). We will assume without loss o generality that v d+1 > since [ v [ v d+1 and v v d+1 are sampled with equal probability and both yield the same sphere. We need only check that S is the orthogonal projection o Π (R d ) H to R d. The equation or Π (R d ) is p c p d+1 = p c =, c d+1 c rv d+1 and the equation or H can be rewritten as p d+1 = v (c p) + 1 cd+1 c v d+1 = v (c p) + 1 v d+1 rv d+1. So, or a point p = [ p p d+1 in the intersection, p c = v (c p) + 1 rv d+1 v d+1 rv d+1.
4 5 th Canadian Conerence on Computational Geometry, 13 Multiplying by and completing the square twice gives p (c rv) = r v d+1 + rv = r, which is precisely the equation or S. Now, we show that the ocal point is a centerpoint o Π (P ). There are several dierent ways to show this, but one nice approach is to observe that or any hyperplane normal to v passing through there is a corresponding hyperplane H normal to [ c rv 1 passing through c. The hyperplane H intersects the paraboloid Φ = { [ p p d+1 pd+1 = p } in an ellipse that projects orthogonally to the very same sphere o radius r centered at c rv. It will suice to show or any p = [ p p H Φ, that p (c rv) = r. By the deinition o H, (c rv) p 1 p = (c rv) c 1 c d+1. Multiplying by and completing the square yields It is tangent to [ R d at the origin, has major radius R and all minor radii all equal to R 1. The stereographic map Π R through [ R, the north pole o this ellipse, rom R d to E R is well-deined as is the inverse map (except at [ R ). As R goes to ininity, E R converges to the paraboloid p d+1 = p /4. This is perhaps easier to see when one observes that E R is the set o points equidistant rom and the sphere centered at [ R 1 with radius R. As R goes to ininity, this sphere becomes a plane and the paraboloid is the set o points equidistant rom a point and a plane. See Figure 1. I we intersect the ellipsoid E R with a hyperplane through one i its ocal points and then stereographically project that intersection to R d rom the pole o the ellipse, then the result is a sphere. The ollowing theorem says that or a ixed hyperplane, it doesn t matter which ellipsoid E R we started with, we always get the same sphere as illustrated in Figure. p (c rv) = c rv (c rv) c + c d+1 = r v c + c d+1 = r. The last equality ollows rom the deinition o r and the act that v = 1. This implies that is a centerpoint o the points Π (P ) because every hyperplane passing through separates the same set o points as the corresponding hy- perplane through c, which is a centerpoint by deinition. Figure : The stereographic projection o the intersection o the ellipse and a plane through the ocal point is the same as the second ocal point is moved up to ininity. Figure 1: Three examples o E R are illustrated or R = 1, R =, and R = rom let to right. In each case, the stereographic map o a single point is illustrated. For the last case, the pole is at ininity, so the result is a vertical projection. 4 Equivalence o Stereographic Maps Let = [ 1. Let R be a real number and consider the ellipsoid E R R d+1 deined as the points p = [ p p d+1 such that p R 1 + (p d+1 R) R = 1. Theorem 3 Let α and β be real numbers in [1, and let R d+1 be any point. I H is a hyperplane containing, then (Π α ) 1 (Π α (Rd ) H) = (Π β ) 1 (Π β (Rd ) H). Proo. The sphere S α = (Π α ) 1 (Π α (Rd ) H) can also be written as S α = {p R d v (Π α (p) ) = }, where v is the normal vector o H. By translating the points in R d and scaling the space uniormly, we may assume that = [ 1. Let R be any real number in the range [1, and let q R d be any point. We will show that Π R (q) lies on a line through that does not
5 CCCG 13, Waterloo, Ontario, August 8 1, 13 depend on R. Thus, v (Π α (q) ) = i and only i v (Π β (q) ) = and thereore, the spheres Sα and S β must be equal or all values o α and β. Let p = [ p p d+1 be the projection Π R(q). All o the relevant points,, p, and q are contained in a plane, so we proceed in two dimensions, letting x = p, y = p d+1, and a = q. x Since p lies on E R, we have R 1 + (y R) R = 1, or equivalently, x R + y(y R)(R 1) =. (1) Since p is also on the line rom [ R, the north pole o E R, to [ q, its planar coordinates x and y satisy the equation y = R a x + R. It ollows that R = ay a(y 1) + x, R 1 =, (a x) a x and y R = xy a x Plugging these values into (1) gives the ollowing. x (ay) y( xy)(a(y 1) + x) + 4(a x) (a x) =. Multiplying by 4(a x) xy and collecting terms yields the line (a 4)x 4a(y 1) =. We observe that this is the equation o a line through = (, 1) as desired. 5 Concluding Remarks The main story o this paper is not entirely new to computational geometry. When the parabolic liting map was irst introduced in the problem o computing Voronoi diagrams by Edelsbrunner and Seidel [4, they replaced previous methods based on the stereographic map. Today, the parabolic liting is the preerred method or reducing Delaunay triangulations to convex hulls in one higher dimension. Throughout, we have worked directly with the Euclidean coordinates. This gave the beneit o making the computations more immediately clear, but came at the cost o considering several special cases at ininity. An alternative approach would be to exploit the power o projective geometry, which has been shown to be the natural language or stereographic projections. Even the parabola is just an ellipse in the projective plane. For more on projective geometry, I highly recommend the book by Richter-Gebert [17. I would like to thank Marc Glisse or helpul conversations and advice on high school algebra. Reerences [1 D. K. Blandord, G. E. Blelloch, D. E. Cardoze, and C. Kadow. Compact representations o simplicial meshes in two and three dimensions. Int. J. Comput. Geometry Appl., 15(1):3 4, 5. [ T. M. Chan. Polynomial-time approximation schemes or packing and piercing at objects. J. Algorithms, 46(): , 3. [3 K. Clarkson, D. Eppstein, G. Miller, C. Sturtivant, and S.-H. Teng. Approximating center points with iterated Radon points. International Journal o Computational Geometry and Applications, 6(3): , Sep invited submission. [4 H. Edelsbrunner and R. Seidel. Voronoi diagrams and arrangements. Discrete & Computational Geometry, 1(1):5 44, [5 D. Eppstein, G. L. Miller, and S.-H. Teng. A deterministic linear time algorithm or geometric separators and its applications. Fundamenta Inormatica, (4):39 39, [6 S. Har-Peled. A simple proo o the existence o a planar separator, July [7 R. J. Lipton, D. J. Rose, and R. E. Tarjan. Generalized nested dissection. SIAM Journal on Numerical Analysis, 16(): , [8 R. J. Lipton and R. E. Tarjan. A separator theorem or planar graphs. SIAM J. Appl. Math, 36: , [9 G. L. Miller and D. R. Sheehy. Approximate centerpoints with proos. Computational Geometry: Theory and Applications, 43(8): , 1. [1 G. L. Miller, S.-H. Teng, W. Thurston, and S. A. Vavasis. Automatic mesh partitioning. In A. George, J. Gilbert, and J. Liu, editors, Graphs Theory and Sparse Matrix Computation, The IMA Volumes in Mathematics and its Application, pages Springer-Verlag, Vol 56. [11 G. L. Miller, S.-H. Teng, W. Thurston, and S. A. Vavasis. Separators or sphere-packings and nearest neighborhood graphs. Journal o the ACM, 44(1):1 9, [1 G. L. Miller, S.-H. Teng, W. Thurston, and S. A. Vavasis. Geometric separators or inite-element meshes. SIAM J. Comput., 19(): , [13 G. L. Miller, S.-H. Teng, and S. A. Vavasis. A uniied geometric approach to graph separators. In Proceedings o the 3nd Annual IEEE Symposium on Foundations o Computer Science, pages , [14 G. L. Miller and W. Thurston. Separators in two and three dimensions. In Proceedings o the th Annual ACM Symposium on Theory o Computing, pages ACM, 199. [15 G. L. Miller and S. A. Vavasis. Density graphs and separators. In Proceedings o the Second Annual ACM- SIAM Symposium on Discrete Algorithms, pages ACM-SIAM, 1991.
6 5 th Canadian Conerence on Computational Geometry, 13 [16 W. Mulzer and D. Werner. Approximating Tverberg points in linear time or any ixed dimension. In Proceedings o the 8th ACM Symposium on Computational Geometry, pages 33 31, 1. [17 J. Richter-Gebert. Perspectives on Projective Geometry. Springer, 11. [18 W. D. Smith and N. C. Wormald. Geometric separator theorems and applications. In Proceedings o the 39th Annual IEEE Symposium on Foundations o Computer Science, pages 3 43, 1998.
Discussion of Location-Scale Depth by I. Mizera and C. H. Müller
Discussion of Location-Scale Depth by I. Mizera and C. H. Müller David Eppstein May 21, 2004 Mizera s previous work [13] introduced tangent depth, a powerful method for defining robust statistics. In this
More informationPersistent Homology and Nested Dissection
Persistent Homology and Nested Dissection Don Sheehy University of Connecticut joint work with Michael Kerber and Primoz Skraba A Topological Data Analysis Pipeline A Topological Data Analysis Pipeline
More informationConic Sections. MATH 211, Calculus II. J. Robert Buchanan. Spring Department of Mathematics
Conic Sections MATH 211, Calculus II J. Robert Buchanan Department o Mathematics Spring 2018 Introduction The conic sections include the parabola, the ellipse, and the hyperbola. y y y x x x Parabola A
More informationLifting Transform, Voronoi, Delaunay, Convex Hulls
Lifting Transform, Voronoi, Delaunay, Convex Hulls Subhash Suri Department of Computer Science University of California Santa Barbara, CA 93106 1 Lifting Transform (A combination of Pless notes and my
More informationImproved Results on Geometric Hitting Set Problems
Improved Results on Geometric Hitting Set Problems Nabil H. Mustafa nabil@lums.edu.pk Saurabh Ray saurabh@cs.uni-sb.de Abstract We consider the problem of computing minimum geometric hitting sets in which,
More informationIn what follows, we will focus on Voronoi diagrams in Euclidean space. Later, we will generalize to other distance spaces.
Voronoi Diagrams 4 A city builds a set of post offices, and now needs to determine which houses will be served by which office. It would be wasteful for a postman to go out of their way to make a delivery
More informationA Proposed Approach for Solving Rough Bi-Level. Programming Problems by Genetic Algorithm
Int J Contemp Math Sciences, Vol 6, 0, no 0, 45 465 A Proposed Approach or Solving Rough Bi-Level Programming Problems by Genetic Algorithm M S Osman Department o Basic Science, Higher Technological Institute
More informationbe a polytope. has such a representation iff it contains the origin in its interior. For a generic, sort the inequalities so that
( Shelling (Bruggesser-Mani 1971) and Ranking Let be a polytope. has such a representation iff it contains the origin in its interior. For a generic, sort the inequalities so that. a ranking of vertices
More informationThe Farthest Point Delaunay Triangulation Minimizes Angles
The Farthest Point Delaunay Triangulation Minimizes Angles David Eppstein Department of Information and Computer Science UC Irvine, CA 92717 November 20, 1990 Abstract We show that the planar dual to the
More informationThe Geometry of Carpentry and Joinery
The Geometry of Carpentry and Joinery Pat Morin and Jason Morrison School of Computer Science, Carleton University, 115 Colonel By Drive Ottawa, Ontario, CANADA K1S 5B6 Abstract In this paper we propose
More informationPiecewise polynomial interpolation
Chapter 2 Piecewise polynomial interpolation In ection.6., and in Lab, we learned that it is not a good idea to interpolate unctions by a highorder polynomials at equally spaced points. However, it transpires
More informationVoronoi Diagrams and Delaunay Triangulations. O Rourke, Chapter 5
Voronoi Diagrams and Delaunay Triangulations O Rourke, Chapter 5 Outline Preliminaries Properties and Applications Computing the Delaunay Triangulation Preliminaries Given a function f: R 2 R, the tangent
More informationPacking Two Disks into a Polygonal Environment
Packing Two Disks into a Polygonal Environment Prosenjit Bose, School of Computer Science, Carleton University. E-mail: jit@cs.carleton.ca Pat Morin, School of Computer Science, Carleton University. E-mail:
More informationGenerell Topologi. Richard Williamson. May 6, 2013
Generell Topologi Richard Williamson May 6, Thursday 7th January. Basis o a topological space generating a topology with a speciied basis standard topology on R examples Deinition.. Let (, O) be a topological
More informationLecture 2 September 3
EE 381V: Large Scale Optimization Fall 2012 Lecture 2 September 3 Lecturer: Caramanis & Sanghavi Scribe: Hongbo Si, Qiaoyang Ye 2.1 Overview of the last Lecture The focus of the last lecture was to give
More informationMAPI Computer Vision. Multiple View Geometry
MAPI Computer Vision Multiple View Geometry Geometry o Multiple Views 2- and 3- view geometry p p Kpˆ [ K R t]p Geometry o Multiple Views 2- and 3- view geometry Epipolar Geometry The epipolar geometry
More informationApproximate Centerpoints with Proofs. Don Sheehy and Gary Miller Carnegie Mellon University
Approximate Centerpoints with Proofs Don Sheehy and Gary Miller Carnegie Mellon University Centerpoints Given a set S R d, a centerpoint is a point p R d such that every closed halfspace containing p contains
More information66 III Complexes. R p (r) }.
66 III Complexes III.4 Alpha Complexes In this section, we use a radius constraint to introduce a family of subcomplexes of the Delaunay complex. These complexes are similar to the Čech complexes but differ
More informationDon t just read it; fight it! Ask your own questions, look for your own examples, discover your own proofs. Is the hypothesis necessary?
Don t just read it; fight it! Ask your own questions, look for your own examples, discover your own proofs. Is the hypothesis necessary? Is the converse true? What happens in the classical special case?
More informationES 240: Scientific and Engineering Computation. a function f(x) that can be written as a finite series of power functions like
Polynomial Deinition a unction () that can be written as a inite series o power unctions like n is a polynomial o order n n ( ) = A polynomial is represented by coeicient vector rom highest power. p=[3-5
More informationA GRAPH FROM THE VIEWPOINT OF ALGEBRAIC TOPOLOGY
A GRAPH FROM THE VIEWPOINT OF ALGEBRAIC TOPOLOGY KARL L. STRATOS Abstract. The conventional method of describing a graph as a pair (V, E), where V and E repectively denote the sets of vertices and edges,
More informationChapter 8. Voronoi Diagrams. 8.1 Post Oce Problem
Chapter 8 Voronoi Diagrams 8.1 Post Oce Problem Suppose there are n post oces p 1,... p n in a city. Someone who is located at a position q within the city would like to know which post oce is closest
More informationCenterpoints and Tverberg s Technique
Centerpoints and Tverberg s Technique Abdul Basit abdulbasit@lums.edu.pk Nabil H. Mustafa nabil@lums.edu.pk Sarfraz Raza sarfrazr@lums.edu.pk Saurabh Ray saurabh@mpi-inf.mpg.de Abstract Using a technique
More informationExact adaptive parallel algorithms for data depth problems. Vera Rosta Department of Mathematics and Statistics McGill University, Montreal
Exact adaptive parallel algorithms for data depth problems Vera Rosta Department of Mathematics and Statistics McGill University, Montreal joint work with Komei Fukuda School of Computer Science McGill
More informationGEOMETRIC TOOLS FOR COMPUTER GRAPHICS
GEOMETRIC TOOLS FOR COMPUTER GRAPHICS PHILIP J. SCHNEIDER DAVID H. EBERLY MORGAN KAUFMANN PUBLISHERS A N I M P R I N T O F E L S E V I E R S C I E N C E A M S T E R D A M B O S T O N L O N D O N N E W
More informationSYNTHESIS OF PLANAR MECHANISMS FOR PICK AND PLACE TASKS WITH GUIDING LOCATIONS
Proceedings of the ASME 2013 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference IDETC/CIE 2013 August 4-7, 2013, Portland, Oregon, USA DETC2013-12021
More informationBichromatic Line Segment Intersection Counting in O(n log n) Time
Bichromatic Line Segment Intersection Counting in O(n log n) Time Timothy M. Chan Bryan T. Wilkinson Abstract We give an algorithm for bichromatic line segment intersection counting that runs in O(n log
More informationPlanar Graphs. 1 Graphs and maps. 1.1 Planarity and duality
Planar Graphs In the first half of this book, we consider mostly planar graphs and their geometric representations, mostly in the plane. We start with a survey of basic results on planar graphs. This chapter
More informationLectures in Discrete Differential Geometry 3 Discrete Surfaces
Lectures in Discrete Differential Geometry 3 Discrete Surfaces Etienne Vouga March 19, 2014 1 Triangle Meshes We will now study discrete surfaces and build up a parallel theory of curvature that mimics
More information3. Voronoi Diagrams. 3.1 Definitions & Basic Properties. Examples :
3. Voronoi Diagrams Examples : 1. Fire Observation Towers Imagine a vast forest containing a number of fire observation towers. Each ranger is responsible for extinguishing any fire closer to her tower
More informationUNIT #2 TRANSFORMATIONS OF FUNCTIONS
Name: Date: UNIT # TRANSFORMATIONS OF FUNCTIONS Part I Questions. The quadratic unction ollowing does,, () has a turning point at have a turning point? 7, 3, 5 5, 8. I g 7 3, then at which o the The structure
More informationG 6i try. On the Number of Minimal 1-Steiner Trees* Discrete Comput Geom 12:29-34 (1994)
Discrete Comput Geom 12:29-34 (1994) G 6i try 9 1994 Springer-Verlag New York Inc. On the Number of Minimal 1-Steiner Trees* B. Aronov, 1 M. Bern, 2 and D. Eppstein 3 Computer Science Department, Polytechnic
More informationBrian Hamrick. October 26, 2009
Efficient Computation of Homology Groups of Simplicial Complexes Embedded in Euclidean Space TJHSST Senior Research Project Computer Systems Lab 2009-2010 Brian Hamrick October 26, 2009 1 Abstract Homology
More information521466S Machine Vision Exercise #1 Camera models
52466S Machine Vision Exercise # Camera models. Pinhole camera. The perspective projection equations or a pinhole camera are x n = x c, = y c, where x n = [x n, ] are the normalized image coordinates,
More informationThe Algorithm Design Manual
Steven S. Skiena The Algorithm Design Manual With 72 Figures Includes CD-ROM THE ELECTRONIC LIBRARY OF SCIENCE Contents Preface vii I TECHNIQUES 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 1.10 2 2.1 2.2 2.3
More informationRough Connected Topologized. Approximation Spaces
International Journal o Mathematical Analysis Vol. 8 04 no. 53 69-68 HIARI Ltd www.m-hikari.com http://dx.doi.org/0.988/ijma.04.4038 Rough Connected Topologized Approximation Spaces M. J. Iqelan Department
More informationTesting Bipartiteness of Geometric Intersection Graphs David Eppstein
Testing Bipartiteness of Geometric Intersection Graphs David Eppstein Univ. of California, Irvine School of Information and Computer Science Intersection Graphs Given arrangement of geometric objects,
More informationOther Voronoi/Delaunay Structures
Other Voronoi/Delaunay Structures Overview Alpha hulls (a subset of Delaunay graph) Extension of Voronoi Diagrams Convex Hull What is it good for? The bounding region of a point set Not so good for describing
More informationCS485/685 Computer Vision Spring 2012 Dr. George Bebis Programming Assignment 2 Due Date: 3/27/2012
CS8/68 Computer Vision Spring 0 Dr. George Bebis Programming Assignment Due Date: /7/0 In this assignment, you will implement an algorithm or normalizing ace image using SVD. Face normalization is a required
More informationPreferred directions for resolving the non-uniqueness of Delaunay triangulations
Preferred directions for resolving the non-uniqueness of Delaunay triangulations Christopher Dyken and Michael S. Floater Abstract: This note proposes a simple rule to determine a unique triangulation
More informationarxiv: v1 [math.co] 25 Sep 2015
A BASIS FOR SLICING BIRKHOFF POLYTOPES TREVOR GLYNN arxiv:1509.07597v1 [math.co] 25 Sep 2015 Abstract. We present a change of basis that may allow more efficient calculation of the volumes of Birkhoff
More informationAlgebraic Geometry of Segmentation and Tracking
Ma191b Winter 2017 Geometry of Neuroscience Geometry of lines in 3-space and Segmentation and Tracking This lecture is based on the papers: Reference: Marco Pellegrini, Ray shooting and lines in space.
More informationQuadratic and cubic b-splines by generalizing higher-order voronoi diagrams
Quadratic and cubic b-splines by generalizing higher-order voronoi diagrams Yuanxin Liu and Jack Snoeyink Joshua Levine April 18, 2007 Computer Science and Engineering, The Ohio State University 1 / 24
More informationarxiv: v1 [cs.cg] 14 Oct 2014
Randomized Triangle Algorithms for Convex Hull Membership Bahman Kalantari Department of Computer Science, Rutgers University, NJ kalantari@cs.rutgers.edu arxiv:1410.3564v1 [cs.cg] 14 Oct 2014 Abstract
More informationComputational Geometry
Lecture 1: Introduction and convex hulls Geometry: points, lines,... Geometric objects Geometric relations Combinatorial complexity Computational geometry Plane (two-dimensional), R 2 Space (three-dimensional),
More informationGeometry and Gravitation
Chapter 15 Geometry and Gravitation 15.1 Introduction to Geometry Geometry is one of the oldest branches of mathematics, competing with number theory for historical primacy. Like all good science, its
More informationComputational Geometry
Computational Geometry 600.658 Convexity A set S is convex if for any two points p, q S the line segment pq S. S p S q Not convex Convex? Convexity A set S is convex if it is the intersection of (possibly
More information1 Proximity via Graph Spanners
CS273: Algorithms for Structure Handout # 11 and Motion in Biology Stanford University Tuesday, 4 May 2003 Lecture #11: 4 May 2004 Topics: Proximity via Graph Spanners Geometric Models of Molecules, I
More informationReflection and Refraction
Relection and Reraction Object To determine ocal lengths o lenses and mirrors and to determine the index o reraction o glass. Apparatus Lenses, optical bench, mirrors, light source, screen, plastic or
More informationA note on Baker s algorithm
A note on Baker s algorithm Iyad A. Kanj, Ljubomir Perković School of CTI, DePaul University, 243 S. Wabash Avenue, Chicago, IL 60604-2301. Abstract We present a corrected version of Baker s algorithm
More informationThe Square Root Phenomenon in Planar Graphs
1 The Square Root Phenomenon in Planar Graphs Survey and New Results Dániel Marx Institute for Computer Science and Control, Hungarian Academy of Sciences (MTA SZTAKI) Budapest, Hungary Satisfiability
More informationConvex Geometry arising in Optimization
Convex Geometry arising in Optimization Jesús A. De Loera University of California, Davis Berlin Mathematical School Summer 2015 WHAT IS THIS COURSE ABOUT? Combinatorial Convexity and Optimization PLAN
More informationOptimal Möbius Transformation for Information Visualization and Meshing
Optimal Möbius Transformation for Information Visualization and Meshing Marshall Bern Xerox Palo Alto Research Ctr. David Eppstein Univ. of California, Irvine Dept. of Information and Computer Science
More informationAlternating Projections
Alternating Projections Stephen Boyd and Jon Dattorro EE392o, Stanford University Autumn, 2003 1 Alternating projection algorithm Alternating projections is a very simple algorithm for computing a point
More informationGeometric structures on manifolds
CHAPTER 3 Geometric structures on manifolds In this chapter, we give our first examples of hyperbolic manifolds, combining ideas from the previous two chapters. 3.1. Geometric structures 3.1.1. Introductory
More information(Master Course) Mohammad Farshi Department of Computer Science, Yazd University. Yazd Univ. Computational Geometry.
1 / 17 (Master Course) Mohammad Farshi Department of Computer Science, Yazd University 1392-1 2 / 17 : Mark de Berg, Otfried Cheong, Marc van Kreveld, Mark Overmars, Algorithms and Applications, 3rd Edition,
More informationRestricted-Orientation Convexity in Higher-Dimensional Spaces
Restricted-Orientation Convexity in Higher-Dimensional Spaces ABSTRACT Eugene Fink Derick Wood University of Waterloo, Waterloo, Ont, Canada N2L3G1 {efink, dwood}@violetwaterlooedu A restricted-oriented
More informationConvex Optimization. Convex Sets. ENSAE: Optimisation 1/24
Convex Optimization Convex Sets ENSAE: Optimisation 1/24 Today affine and convex sets some important examples operations that preserve convexity generalized inequalities separating and supporting hyperplanes
More informationTesting Isomorphism of Strongly Regular Graphs
Spectral Graph Theory Lecture 9 Testing Isomorphism of Strongly Regular Graphs Daniel A. Spielman September 26, 2018 9.1 Introduction In the last lecture we saw how to test isomorphism of graphs in which
More informationLower Bounds for Geometric Diameter Problems
Lower Bounds for Geometric Diameter Problems p. 1/48 Lower Bounds for Geometric Diameter Problems Hervé Fournier University of Versailles St-Quentin en Yvelines Antoine Vigneron INRA Jouy-en-Josas Lower
More information60 2 Convex sets. {x a T x b} {x ã T x b}
60 2 Convex sets Exercises Definition of convexity 21 Let C R n be a convex set, with x 1,, x k C, and let θ 1,, θ k R satisfy θ i 0, θ 1 + + θ k = 1 Show that θ 1x 1 + + θ k x k C (The definition of convexity
More informationGeometric Unique Set Cover on Unit Disks and Unit Squares
CCCG 2016, Vancouver, British Columbia, August 3 5, 2016 Geometric Unique Set Cover on Unit Disks and Unit Squares Saeed Mehrabi Abstract We study the Unique Set Cover problem on unit disks and unit squares.
More informationWHEN DO THREE LONGEST PATHS HAVE A COMMON VERTEX?
WHEN DO THREE LONGEST PATHS HAVE A COMMON VERTEX? MARIA AXENOVICH Abstract. It is well known that any two longest paths in a connected graph share a vertex. It is also known that there are connected graphs
More informationUnlabeled equivalence for matroids representable over finite fields
Unlabeled equivalence for matroids representable over finite fields November 16, 2012 S. R. Kingan Department of Mathematics Brooklyn College, City University of New York 2900 Bedford Avenue Brooklyn,
More informationMath-3 Lesson 1-7 Analyzing the Graphs of Functions
Math- Lesson -7 Analyzing the Graphs o Functions Which unctions are symmetric about the y-axis? cosx sin x x We call unctions that are symmetric about the y -axis, even unctions. Which transormation is
More informationMATRIX ALGORITHM OF SOLVING GRAPH CUTTING PROBLEM
UDC 681.3.06 MATRIX ALGORITHM OF SOLVING GRAPH CUTTING PROBLEM V.K. Pogrebnoy TPU Institute «Cybernetic centre» E-mail: vk@ad.cctpu.edu.ru Matrix algorithm o solving graph cutting problem has been suggested.
More informationThe Graph of an Equation Graph the following by using a table of values and plotting points.
Calculus Preparation - Section 1 Graphs and Models Success in math as well as Calculus is to use a multiple perspective -- graphical, analytical, and numerical. Thanks to Rene Descartes we can represent
More informationMath 451: Euclidean and Non-Euclidean Geometry MWF 3pm, Gasson 204 Homework 8 Solutions
Math 451: Euclidean and Non-Euclidean Geometry MWF 3pm, Gasson 204 Homework 8 Solutions Exercises from Chapter 2: 5.5, 5.10, 5.13, 5.14 Exercises from Chapter 3: 1.2, 1.3, 1.5 Exercise 5.5. Give an example
More informationApproximating Polygonal Objects by Deformable Smooth Surfaces
Approximating Polygonal Objects by Deformable Smooth Surfaces Ho-lun Cheng and Tony Tan School of Computing, National University of Singapore hcheng,tantony@comp.nus.edu.sg Abstract. We propose a method
More informationREVIEW OF FUZZY SETS
REVIEW OF FUZZY SETS CONNER HANSEN 1. Introduction L. A. Zadeh s paper Fuzzy Sets* [1] introduces the concept of a fuzzy set, provides definitions for various fuzzy set operations, and proves several properties
More informationThe Unified Segment Tree and its Application to the Rectangle Intersection Problem
CCCG 2013, Waterloo, Ontario, August 10, 2013 The Unified Segment Tree and its Application to the Rectangle Intersection Problem David P. Wagner Abstract In this paper we introduce a variation on the multidimensional
More information1 Appendix to notes 2, on Hyperbolic geometry:
1230, notes 3 1 Appendix to notes 2, on Hyperbolic geometry: The axioms of hyperbolic geometry are axioms 1-4 of Euclid, plus an alternative to axiom 5: Axiom 5-h: Given a line l and a point p not on l,
More information6. Concluding Remarks
[8] K. J. Supowit, The relative neighborhood graph with an application to minimum spanning trees, Tech. Rept., Department of Computer Science, University of Illinois, Urbana-Champaign, August 1980, also
More informationGenerating Tool Paths for Free-Form Pocket Machining Using z-buffer-based Voronoi Diagrams
Int J Adv Manuf Technol (1999) 15:182 187 1999 Springer-Verlag London Limited Generating Tool Paths for Free-Form Pocket Machining Using z-buffer-based Voronoi Diagrams Jaehun Jeong and Kwangsoo Kim Department
More informationSurface Reconstruction. Gianpaolo Palma
Surface Reconstruction Gianpaolo Palma Surface reconstruction Input Point cloud With or without normals Examples: multi-view stereo, union of range scan vertices Range scans Each scan is a triangular mesh
More informationApplications of Geometric Spanner
Title: Name: Affil./Addr. 1: Affil./Addr. 2: Affil./Addr. 3: Keywords: SumOriWork: Applications of Geometric Spanner Networks Joachim Gudmundsson 1, Giri Narasimhan 2, Michiel Smid 3 School of Information
More informationDifferential Geometry: Circle Patterns (Part 1) [Discrete Conformal Mappinngs via Circle Patterns. Kharevych, Springborn and Schröder]
Differential Geometry: Circle Patterns (Part 1) [Discrete Conformal Mappinngs via Circle Patterns. Kharevych, Springborn and Schröder] Preliminaries Recall: Given a smooth function f:r R, the function
More informationMidpoint Routing algorithms for Delaunay Triangulations
Midpoint Routing algorithms for Delaunay Triangulations Weisheng Si and Albert Y. Zomaya Centre for Distributed and High Performance Computing School of Information Technologies Prologue The practical
More informationChapter 3 Image Enhancement in the Spatial Domain
Chapter 3 Image Enhancement in the Spatial Domain Yinghua He School o Computer Science and Technology Tianjin University Image enhancement approaches Spatial domain image plane itsel Spatial domain methods
More information2. Convex sets. affine and convex sets. some important examples. operations that preserve convexity. generalized inequalities
2. Convex sets Convex Optimization Boyd & Vandenberghe affine and convex sets some important examples operations that preserve convexity generalized inequalities separating and supporting hyperplanes dual
More informationConnected Components of Underlying Graphs of Halving Lines
arxiv:1304.5658v1 [math.co] 20 Apr 2013 Connected Components of Underlying Graphs of Halving Lines Tanya Khovanova MIT November 5, 2018 Abstract Dai Yang MIT In this paper we discuss the connected components
More informationGeometric structures on manifolds
CHAPTER 3 Geometric structures on manifolds In this chapter, we give our first examples of hyperbolic manifolds, combining ideas from the previous two chapters. 3.1. Geometric structures 3.1.1. Introductory
More informationModule 3: Stand Up Conics
MATH55 Module 3: Stand Up Conics Main Math concepts: Conic Sections (i.e. Parabolas, Ellipses, Hyperbolas), nd degree equations Auxilliary ideas: Analytic vs. Co-ordinate-free Geometry, Parameters, Calculus.
More informationPentagon contact representations
Pentagon contact representations Stean Felsner Institut ür Mathematik Technische Universität Berlin Hendrik Schrezenmaier Institut ür Mathematik Technische Universität Berlin August, 017 Raphael Steiner
More informationInversive Plane Geometry
Inversive Plane Geometry An inversive plane is a geometry with three undefined notions: points, circles, and an incidence relation between points and circles, satisfying the following three axioms: (I.1)
More informationTutte Embeddings of Planar Graphs
Spectral Graph Theory and its Applications Lectre 21 Ttte Embeddings o Planar Graphs Lectrer: Daniel A. Spielman November 30, 2004 21.1 Ttte s Theorem We sally think o graphs as being speciied by vertices
More informationDrawing some 4-regular planar graphs with integer edge lengths
CCCG 2013, Waterloo, Ontario, August 8 10, 2013 Drawing some 4-regular planar graphs with integer edge lengths Timothy Sun Abstract A classic result of Fáry states that every planar graph can be drawn
More informationMonotone Paths in Geometric Triangulations
Monotone Paths in Geometric Triangulations Adrian Dumitrescu Ritankar Mandal Csaba D. Tóth November 19, 2017 Abstract (I) We prove that the (maximum) number of monotone paths in a geometric triangulation
More information2. Convex sets. x 1. x 2. affine set: contains the line through any two distinct points in the set
2. Convex sets Convex Optimization Boyd & Vandenberghe affine and convex sets some important examples operations that preserve convexity generalized inequalities separating and supporting hyperplanes dual
More informationThe angle measure at for example the vertex A is denoted by m A, or m BAC.
MT 200 ourse notes on Geometry 5 2. Triangles and congruence of triangles 2.1. asic measurements. Three distinct lines, a, b and c, no two of which are parallel, form a triangle. That is, they divide the
More informationConvexity: an introduction
Convexity: an introduction Geir Dahl CMA, Dept. of Mathematics and Dept. of Informatics University of Oslo 1 / 74 1. Introduction 1. Introduction what is convexity where does it arise main concepts and
More informationSeparators in Two and Three Dimensions
Separators in Two and Three Dimensions. Gary L. Miller* School of Computer Science Carnegie Mellon University Dept of Computer Science University of Southern California William Thurston Dept of Mathematics
More informationComputational Geometry Algorithmische Geometrie
Algorithmische Geometrie Panos Giannopoulos Wolfgang Mulzer Lena Schlipf AG TI SS 2013 !! Register in Campus Management!! Outline What you need to know (before taking this course) What is the course about?
More information1. A model for spherical/elliptic geometry
Math 3329-Uniform Geometries Lecture 13 Let s take an overview of what we know. 1. A model for spherical/elliptic geometry Hilbert s axiom system goes with Euclidean geometry. This is the same geometry
More information6.854J / J Advanced Algorithms Fall 2008
MIT OpenCourseWare http://ocw.mit.edu 6.854J / 18.415J Advanced Algorithms Fall 2008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms. 18.415/6.854 Advanced
More informationMATH 890 HOMEWORK 2 DAVID MEREDITH
MATH 890 HOMEWORK 2 DAVID MEREDITH (1) Suppose P and Q are polyhedra. Then P Q is a polyhedron. Moreover if P and Q are polytopes then P Q is a polytope. The facets of P Q are either F Q where F is a facet
More informationGeometric Red-Blue Set Cover for Unit Squares and Related Problems
Geometric Red-Blue Set Cover for Unit Squares and Related Problems Timothy M. Chan Nan Hu December 1, 2014 Abstract We study a geometric version of the Red-Blue Set Cover problem originally proposed by
More informationOutline of the presentation
Surface Reconstruction Petra Surynková Charles University in Prague Faculty of Mathematics and Physics petra.surynkova@mff.cuni.cz Outline of the presentation My work up to now Surfaces of Building Practice
More informationHausdorff Approximation of 3D Convex Polytopes
Hausdorff Approximation of 3D Convex Polytopes Mario A. Lopez Department of Mathematics University of Denver Denver, CO 80208, U.S.A. mlopez@cs.du.edu Shlomo Reisner Department of Mathematics University
More informationPlanar graphs. Math Prof. Kindred - Lecture 16 Page 1
Planar graphs Typically a drawing of a graph is simply a notational shorthand or a more visual way to capture the structure of the graph. Now we focus on the drawings themselves. Definition A drawing of
More information