Surfaces of finite types Chihchy Fwu Department of Mathematics Soochow University November 2001.

Slides:



Advertisements
Similar presentations
CS 691 Computational Photography Instructor: Gianfranco Doretto Image Warping.
Advertisements

Stability and its Ramifications M.S. Narasimhan 1.
Input Space versus Feature Space in Kernel- Based Methods Scholkopf, Mika, Burges, Knirsch, Muller, Ratsch, Smola presented by: Joe Drish Department of.
Computational Topology for Computer Graphics Klein bottle.
Surface Classification Using Conformal Structures Xianfeng Gu 1, Shing-Tung Yau 2 1. Computer and Information Science and Engineering, University of Florida.
8 CHAPTER Linear Algebra: Matrix Eigenvalue Problems Chapter 8 p1.
Signal , Weight Vector Spaces and Linear Transformations
6. One-Dimensional Continuous Groups 6.1 The Rotation Group SO(2) 6.2 The Generator of SO(2) 6.3 Irreducible Representations of SO(2) 6.4 Invariant Integration.
Image Warping : Computational Photography Alexei Efros, CMU, Fall 2008 Some slides from Steve Seitz
Variational Principles and Rigidity on Triangulated Surfaces Feng Luo Rutgers University Geometry & Topology Down Under The Hyamfest Melbourne, Australia.
Implicit Surfaces Tom Ouyang January 29, Outline Properties of Implicit Surfaces Polygonization Ways of generating implicit surfaces Applications.
Audrey Terras CRM Montreal 2009 Lecture 1: Riemann, Dedekind, Selberg, and Ihara Zetas.
Geometric Modeling Surfaces Mortenson Chapter 6 and Angel Chapter 9.
6 1 Linear Transformations. 6 2 Hopfield Network Questions.
17 VECTOR CALCULUS.
Lecture # 32 (Last) Dr. SOHAIL IQBAL
Boyce/DiPrima 9th ed, Ch 11.2: Sturm-Liouville Boundary Value Problems Elementary Differential Equations and Boundary Value Problems, 9th edition, by.
Multiple Integrals 12. Surface Area Surface Area In this section we apply double integrals to the problem of computing the area of a surface.
1 Dr. Scott Schaefer Generalized Barycentric Coordinates.
Subdivision surfaces Construction and analysis Martin Reimers CMA/IFI, University of Oslo September 24th 2004.
Quantum Mechanics(14/2)Taehwang Son Functions as vectors  In order to deal with in more complex problems, we need to introduce linear algebra. Wave function.
Mathematics for Computer Graphics (Appendix A) Won-Ki Jeong.
Signal Processing and Representation Theory Lecture 1.
TEMPLATE BASED SHAPE DESCRIPTOR Raif Rustamov Department of Mathematics and Computer Science Drew University, Madison, NJ, USA.
1 Dr. Scott Schaefer Generalized Barycentric Coordinates.
Cindy Grimm Parameterization with Manifolds Cindy Grimm.
6 1 Linear Transformations. 6 2 Hopfield Network Questions The network output is repeatedly multiplied by the weight matrix W. What is the effect of this.
1 Surface Applications Fitting Manifold Surfaces To 3D Point Clouds, Cindy Grimm, David Laidlaw and Joseph Crisco. Journal of Biomechanical Engineering,
1 Adding charts anywhere Assume a cow is a sphere Cindy Grimm and John Hughes, “Parameterizing n-holed tori”, Mathematics of Surfaces X, 2003 Cindy Grimm,
Polynomial Preserving Gradient Recovery in Finite Element Methods Zhimin Zhang Department of Mathematics Wayne State University Detroit, MI 48202
Computer Graphics Some slides courtesy of Pierre Alliez and Craig Gotsman Texture mapping and parameterization.
6. Introduction to Spectral method. Finite difference method – approximate a function locally using lower order interpolating polynomials. Spectral method.
Signal Processing and Representation Theory Lecture 2.
§ Linear Operators Christopher Crawford PHY
16/5/ :47 UML Computer Graphics Conceptual Model Application Model Application Program Graphics System Output Devices Input Devices API Function.
Signal Processing and Representation Theory Lecture 3.
SVD Data Compression: Application to 3D MHD Magnetic Field Data Diego del-Castillo-Negrete Steve Hirshman Ed d’Azevedo ORNL ORNL-PPPL LDRD Meeting ORNL.
GRASP Learning a Kernel Matrix for Nonlinear Dimensionality Reduction Kilian Q. Weinberger, Fei Sha and Lawrence K. Saul ICML’04 Department of Computer.
Climate Modeling In-Class Discussion: Legendre Polynomials.
1 The Mathematics of Quantum Mechanics 3. State vector, Orthogonality, and Scalar Product.
Image transformations Digital Image Processing Instructor: Dr. Cheng-Chien LiuCheng-Chien Liu Department of Earth Sciences National Cheng Kung University.
Geometry of Shape Manifolds
Tony Jebara, Columbia University Advanced Machine Learning & Perception Instructor: Tony Jebara.
Positively Expansive Maps and Resolution of Singularities Wayne Lawton Department of Mathematics National University of Singapore
Copyright © Cengage Learning. All rights reserved. 16 Vector Calculus.
Numerical methods 1 An Introduction to Numerical Methods For Weather Prediction by Mariano Hortal office 122.
11/6/ :55 Graphics II Introduction to Parametric Curves and Surfaces Session 2.
CSCI 425/ D Mathematical Preliminaries. CSCI 425/525 2 Coordinate Systems Z X Y Y X Z Right-handed coordinate system Left-handed coordinate system.
Signal & Weight Vector Spaces
Parametric Surfaces and their Area Part I
Value Function Approximation with Diffusion Wavelets and Laplacian Eigenfunctions by S. Mahadevan & M. Maggioni Discussion led by Qi An ECE, Duke University.
Euler characteristic (simple form):
Function Representation & Spherical Harmonics. Function approximation G(x)... function to represent B 1 (x), B 2 (x), … B n (x) … basis functions G(x)
Principal Warps: Thin-Plate Splines and the Decomposition of Deformations 김진욱 ( 이동통신망연구실 ; 박천현 (3D 모델링 및 처리연구실 ;
Reduced echelon form Matrix equations Null space Range Determinant Invertibility Similar matrices Eigenvalues Eigenvectors Diagonabilty Power.
Beyond Vectors Hung-yi Lee. Introduction Many things can be considered as “vectors”. E.g. a function can be regarded as a vector We can apply the concept.
Image Transformation Spatial domain (SD) and Frequency domain (FD)
Spectral partitioning works: Planar graphs and finite element meshes
Introduction The central problems of Linear Algebra are to study the properties of matrices and to investigate the solutions of systems of linear equations.
Introduction The central problems of Linear Algebra are to study the properties of matrices and to investigate the solutions of systems of linear equations.
Intrinsic Data Geometry from a Training Set
Copyright © Cengage Learning. All rights reserved.
PHY 711 Classical Mechanics and Mathematical Methods
Elementary Linear Algebra
PHY 711 Classical Mechanics and Mathematical Methods
PHY 711 Classical Mechanics and Mathematical Methods
MA5242 Wavelets Lecture 1 Numbers and Vector Spaces
A Portrait of a Group on a Surface with Boundary
Computer Aided Geometric Design
Presentation transcript:

Surfaces of finite types Chihchy Fwu Department of Mathematics Soochow University November 2001

How to visualize abstractly defined surfaces?

What is the configuration space of the following linkage [W. Thurston 1982]

It is a surface of genus 3 Vertice = 8  6 / 4 = 12 Edges = 8  6 / 2 = 24 Faces = 8  = 12 – = 2 ( 1 – g )

Outline Review classical surfaces Type number Computation application

Circle Group SO(2) X =  cos t sin t   -sin t cos t  X ” + X = 0

Orthogonal group SO(n) X = ( x ij )  R n  n Killing form ds 2 = - tr( dX dX t ) Orthonormal basis of Lie algebra E ij = e i e j t – e j e i t Casimir operator  E ij 2 = - I Maurer Cartan form  = dX X t Structure equation d  =    Laplacian  X + X = 0

Spectrum of a Manifold Eigenfunction  with eigenvalue   +  = 0 Eigenvalues 0 < 1 < 2 < 3 < …. E( k ) space of eigenfunctions is a finite dimensional vector space

Immersion of homogenous space M = G / H Riemannian homogeneous space G compact Lie group H acts on T [H] M irreducibly. {  1,  2, …,  N } an orthonormal basis of E( k ) The evaluation map  : M  R N  (x) = (  1 (x),  2 (x), …,  N (x) ) is an isometric immersion and   = k .

Some examples Sphere S n = SO(n+1) / SO(n) S 2n+1 = SU(n+1) / SU(n) S 4n+3 = Sp(n+1) / Sp(n) Projective space RP n = SO(n+1) / S(O(n)  O(1)) CP n = SU(n+1) / S(U(n)  U(1)) QP n = Sp(n+1) / Sp(n)  Sp(1)

spherical harmonics E( k ) consists of harmonic polynomials homogeneous of degree k in x, y, z dim E( k ) = 2 k +1 k = k ( k + 1)

E( 1 ) = { x, y, z } canonical immersion

E( 2 ) = { x 2 - y 2, y 2 - z 2, z 2 - x 2, 2xy, 2yz, 2zx } Veronese surface as an immersion of the real projective plane RP 2  R 5  S 5 = S 4  R 4

Roman Surface ( 2xy, 2yz, 2zx )

Crosscap ( 2yz, 2zx, z 2 -x 2 )

E( 3 ) Immersion of the 3rd degree

E( 4 ) Boy surface immersion of degree 4

Boy surface viewed from bottom

Boy surface viewed from top

3 different views of Boy surface

Harmonic immersions of Flat torus T 2 (a,b)   S 1 ( a ) × S 1 ( b )  S 3  R 4, a 2 + b 2 = 1  (u,v) = ( a cos u, a sin u, b cos v, b sin v ) The harmonic functions on a flat torus are ordinary trig polynomials

Anchor ring = Flat torus under stereographic projection

Flat torus under spherical harmonic immersion of degree 2

Another spherical harmonic immersion of Flat torus

Type number X : M n  R N is a Riemannian manifold. E( k ) is the space of kth eigenfunctions of the Laplacian. The position vector has a Fourier expansion X =  X k Type number = # { k | X k  0 } All previous examples have type number either 1 : homogeneous spaces, or 2 : flat torus

QUESTION Is type number necessarily finite ? What is the type number of a surface ?

Spectral problem Eigenfunction  with eigenvalue   +  = 0 Rayleigh quotient Q(  ) = |  | 2 / |  | 2

Rayleigh-Ritz Approximation parametrize the surface in a single patch  = I  I with boundary properly identified. subdivide  into cells {e k } by a grid {x k } define a finite element  k at each vertex x k approximate the function  by  u k  k with coefficients vector u=(u k ) k ≦ K set K = 1600

Generalized eigenvalue problem Replace the Rayleigh quotient with the discrete form  Au, u  /  Bu, u  SVD gives A U = B U D U = ( u 1 … u K ) the matrix of eigenvectors, and D = diag( 1 … K ) the matrix of eigenvalues A u = B u

Determine the type number discretized position vector X = U Y Y = ( Y 1,Y 2, Y 3 ) Fourier coefficients y = Y 1 2 +Y Y 3 2 = ( y k ) Type number = # { k | y k > residue} Here residue = 10 -6

Surfaces to be investigated sphere flat torus a class of anchor rings a class of that under inversion a class of knotted tori a surface of genus two

Reconstructed sphere

Reconstructed Roman surface

Reconstructed flat torus

Reconstructed flat torus with higher degree eigenfunctions

Class of anchor rings

Generators are the geodesic circles at (1,0) of H 2

Type numbers of anchor rings

Class of inversions T 2 (a,b)  S 3  S 3   R 3  R 3 O(1,4) is the conformal groups of both R 3 and S 3 Inversion : x, u  R 4, | x | = 1  ( x ) = u – ( |u| 2 –1) / | x- u| 2 (x - u )

Inversion of an anchor ring

Type numbers of inversions

Class of knotted tori tube around a torus knot

Type numbers of knotted tori

Surface of genus 2 type number > 1525

Surface of genus 3 type number = ?

Surface of genus 5 type number = ?

All surfaces are of finite type, but some are more finite than the others [rephase George Orwell]

Surface of higher genus Uniformization : a surface M g of genus g > 1 is covered by the unit disk  M g =  \ SL(2, R) / SO(2)  =  1 ( M g ) fundamental group =  a 1, b 1,… a g, b g | [a 1,b 1 ][a 2,b 2 ]…[a g,b g ] = 1  M g  CP 3  R N

Fundamental domain ( g = 2)

Riemann surface of genus 2

View from bottom

View from top

謝謝聽講