Voronoi Diagram (Supplemental)

40
Voronoi Diagram (Supplem ental) The Universal Spatial Dat a Structure (Franz Aurenh ammer)

description

Voronoi Diagram (Supplemental). The Universal Spatial Data Structure (Franz Aurenhammer ). Outline. Voronoi and Delaunay Facility location problem Nearest neighbor Fortune ’ s algorithm revisited Generalized Voronoi diagrams. Voronoi Diagram. Dual: Delaunay Triangulation. - PowerPoint PPT Presentation

Transcript of Voronoi Diagram (Supplemental)

Page 1: Voronoi Diagram (Supplemental)

Voronoi Diagram (Supplemental)

The Universal Spatial Data Structure (Franz Aurenhammer)

Page 2: Voronoi Diagram (Supplemental)

Fall 2005 2

Outline

Voronoi and DelaunayFacility location problemNearest neighborFortune’s algorithm revisitedGeneralized Voronoi diagrams

Page 3: Voronoi Diagram (Supplemental)

Fall 2005 3

Voronoi Diagram

Page 4: Voronoi Diagram (Supplemental)

Fall 2005 4

Dual: Delaunay Triangulation

Page 5: Voronoi Diagram (Supplemental)

Fall 2005 5

Facility Location Problems

Determine a location to minimize the distance to its furthest customerMinimum enclosing circle

Determine a location whose distance to nearest store is as large as possibleLargest empty circle

ip

jp

kp

q

Page 6: Voronoi Diagram (Supplemental)

Fall 2005 6

Facility Location (version 2)

Seek location for new grocery store, whose distance to nearest store is as large as possible — center of largest empty circleOne restriction: center in convex hull of the sites

Page 7: Voronoi Diagram (Supplemental)

Fall 2005 7

Facility Location (cont)

Center in hull: p must be coincident with a voronoi vertex

Center on hull: p must lie on a voronoi edge

Page 8: Voronoi Diagram (Supplemental)

Fall 2005 8

Largest Empty Circle

Page 9: Voronoi Diagram (Supplemental)

Fall 2005 9

Nearest Neighbor Search

A special case of point-location problem where every face in the subdivision is monotoneUse chain method to get O(log n) time complexity for query

Page 10: Voronoi Diagram (Supplemental)

Fall 2005 10

Page 11: Voronoi Diagram (Supplemental)

Fall 2005 11

1

2

34

56

8

7

Page 12: Voronoi Diagram (Supplemental)

Fall 2005 12

1

2

34

56

8

7

Page 13: Voronoi Diagram (Supplemental)

Fall 2005 13

1

2

34

56

8

7

Page 14: Voronoi Diagram (Supplemental)

Fall 2005 14

1

2

34

56

8

7

Page 15: Voronoi Diagram (Supplemental)

Fall 2005 15

1

2

34

56

8

7

Page 16: Voronoi Diagram (Supplemental)

Fall 2005 16

1

2

34

56

8

7

Page 17: Voronoi Diagram (Supplemental)

Fall 2005 17

1

2

34

56

8

7

Page 18: Voronoi Diagram (Supplemental)

Fall 2005 18

Cluster Analysis

Page 19: Voronoi Diagram (Supplemental)

Fall 2005 19

Closest Pairs

In collision detection, two closest sites are in greatest danger of collisionNaïve approach: (n2)

Each site and its closest pair share an edge check all Voronoi edges O(n)Furthest pair cannot be derived directly from the diagram

Page 20: Voronoi Diagram (Supplemental)

Fall 2005 20

Motion Planning (translational)

Collision avoidance:

stay away from obstacle

Page 21: Voronoi Diagram (Supplemental)

Fall 2005 21

Fortune’s Algorithm Revisited

ConesIdeaH/W implementation

The curve of intersection of two cones projects to a line.

Page 22: Voronoi Diagram (Supplemental)

Fall 2005 22

45 deg Cone

distance=heightsite

Page 23: Voronoi Diagram (Supplemental)

Fall 2005 23

Cone (cont)

intersection of cone equal-distance point

Page 24: Voronoi Diagram (Supplemental)

Fall 2005 24

Cone (cont)When viewed from –Z, we got colored V-cells

Page 25: Voronoi Diagram (Supplemental)

Fall 2005 25

Nearest Distance Function

Viewed from here[less than]

Page 26: Voronoi Diagram (Supplemental)

Fall 2005 26

Furthest Distance Function

Viewed from here[greater than]

Page 27: Voronoi Diagram (Supplemental)

Fall 2005 27

Fortune’s Algorithm (Cont)

Cone slicing

Cone cut up by sweep plane and L are sweeping toward the right.

Page 28: Voronoi Diagram (Supplemental)

Fall 2005 28

Fortune’s Algorithm (Cont)

Viewed from z = -, The heavy curve is the parabolic front.

How the 2D algorithm and

the 3D cones are related…

Page 29: Voronoi Diagram (Supplemental)

Fall 2005 29

Generalized Voronoi Diagram

V(points), Euclidean distance V(points, lines, curves, …)Distance function: Euclidean, weighted, farthest

Page 30: Voronoi Diagram (Supplemental)

Fall 2005 30

Brute Force Method

Record ID of the closest site to each sample

point

Coarsepoint-sampling

result

Finerpoint-sampling

result

Page 31: Voronoi Diagram (Supplemental)

Fall 2005 31

Graphics Hardware Acceleration

Our 2-part discrete Voronoidiagram representation

Distance

Depth Buffer

Site IDs

Color Buffer

Simply rasterize the cones using gra

phics hardware

Haeberli90, Woo97

Page 32: Voronoi Diagram (Supplemental)

Fall 2005 32

Algorithm

Associate each primitive with the corresponding distance meshRender each distance mesh with depth test onVoronoi edges: found by continuation methods

Page 33: Voronoi Diagram (Supplemental)

Fall 2005 33

Ex: Voronoi diagram between a point and a line

Page 34: Voronoi Diagram (Supplemental)

Fall 2005 34

Distance Meshes

linecurve

polygon

Page 35: Voronoi Diagram (Supplemental)

Fall 2005 35

Applications (Mosaic)

Page 36: Voronoi Diagram (Supplemental)

Fall 2005 36

Hausner01, siggraph

Page 37: Voronoi Diagram (Supplemental)

Fall 2005 37

Medial Axis Computation

Medial axes as part of Voronoi diagram

Page 38: Voronoi Diagram (Supplemental)

Fall 2005 38

Piano Mover: Real-time Motion Planning (static and dynamic)

Plan motion of piano through 100K triangle model

Distance buffer of floorplan used as potential field

Page 39: Voronoi Diagram (Supplemental)

Fall 2005 39

Variety of Voronoi Diagram

(regular) Voronoi diagram

Furthest distance Voronoi diagram

Page 40: Voronoi Diagram (Supplemental)

Fall 2005 40

Minimum Enclosing Circle

Center of MEC is at the vertex of furthest site Voronoi diagram