Superquadric Glyphs for Symmetric Second- Order Tensors
|
|
- Sherilyn Murphy
- 5 years ago
- Views:
Transcription
1 Superquadric Glyphs for Symmetric Second- Order Tensors Thomas Schultz, Gordon L. Kindlmann Computer Science Dept, Computation Institute University of Chicago Symmetric Tensor Representations [Kindlmann 24] R R -1 Eigenvalues tell us about Positive Definiteness
2 Symmetric Tensor Representations Topic of this talk R R -1 Tensor is indefinite Why care about indefinite tensors? Indefinite Symmetric Tensors arise in many fields: Hessians Stress Tensors Rate-of-deformation Tensors Geometry Tensors
3 Principles for Tensor Glyph Design Faithful and expressive visualization requires: Preservation of Symmetry: Glyph should have same symmetries as the tensor Continuity: Disambiguity: Principles for Tensor Glyph Design Natural for a wide range of applications: Invariance under scaling: Invariance under projection to eigenplanes:
4 Glyph Coloring We color each point x on the glyph by sign(x T Dx) + red - blue Satisfies! Preservation of Symmetry! Continuity! Disambiguity Our Fundamental Glyph Equation scaling rotation normalized eigenvalues basis shape
5 Explicit Ellipses Dx for x 1 " Symmetry Preservation " Disambiguity Implicit Ellipses x T D -2 x 1 sign preserving # Unbounded Surfaces
6 Exp-Ellipses exp(d)x for x 1 " Disambiguity Reynolds Glyphs (x T D -2 x)x for x 1 " Invariance under Projection # Occlusions in 3D
7 New Glyph for 2D symmetric tensor Meet disambiguity, symmetry preservation req. Glyph shape shows eigenvalue sign differences Convex indicates same sign (positive-definite, negative-definite) Concave indicates different sign (indefinite) From 2D to 3D tensor glyphs Understand space of 3D symmetric tensor shapes 2D (, ) Scaffold (prototype) 3D glyphs with 2D glyphs λ 2 (+,+) λ 1 (+, ) 3D? Find mapping from 3D tensor shapes to 3D superquadrics
8 Space of 3D tensor shape Eigenvalue sorting!1 "!2 "!3 Lune of tensor shape (+,+)!1 "!2!2 "!3!1 "!2 "!3 λ1 (, ) (+, ) +!3 +!3 +!3 +!1 +!2 +!1 +!2 +!1 +!2 +!1 +!1 +!1 +!2 +!2 +!2 -!3 -!3 -!3 > < λ 1 λ 3 λ 2 λ 3 λ 2 λ 3 λ1 λ3 λ1 λ3 λ 3 λ 2 λ 3 λ 1 λ 2 λ 1 Positive Definite λ 1 λ 2 λ 3 > λ 1 λ 2 λ 3 < Indefinite (mixed eval signs) λ 1 > λ 1 < Negative Definite Space of 3D tensor shape Mercator projection!1!2!3 >!1!2!3 < Kindlmann Superquadric Tensor Glyphs, Vissym 4
9 λ 3 1 λ 3 2 λ λ λ1 λ 2 λ3 λ + 1 λ 2 λ 2 λ1 λ 3 λ 1 λ 2 λ 3 λ3 λ3 +λ1 λ1 λ1 λ3 λ1 > λ1 < λ1 λ1 λ 3 λ1 v 3 λ + λ +λ λ3 λ < λ Cubic projection unfold, shear # unit square of tensor shape 1 λ 1 λ 2 λ 3 λ3 λ3 > λ3 < λ1 > λ3 Space of 3D tensor shape λ1 λ 2 λ3 u λ3 > λ3 λ3 > λ3 < < λ1 3D (+,+)? λ 3 λ1 λ3 λ1 λ 2 Find mapping from 3D tensor shapes to 3D superquadrics λ3 Scaffold (prototype) 3D glyphs with 2D glyphs λ1 (+, ) (, ) λ1 λ 1 λ 2 λ 3 λ1 Understand space of 3D tensor shapes 2D λ 1 λ 2 λ 3 From 2D to 3D tensor glyphs λ1 > λ1 <
10 From 2D to 3D tensor glyphs Understand space of 3D tensor shapes 2D (, ) λ 2 (+,+) λ 1 (+, ) 3D > <? λ1 λ3 λ3 λ3 λ1 λ3 λ1 λ3 λ3 λ3 λ1 λ1 λ1 λ3 > λ1 λ3 < λ1 > λ1 < Scaffold (prototype) 3D glyphs with 2D glyphs Find mapping from 3D tensor shapes to 3D superquadrics Invariance under eigenplane projection! orthogonal 2D glyphs form scaffold for 3D glyphs Can use these 3D shapes to guide next step...
11 From 2D to 3D tensor glyphs Understand space of 3D tensor shapes 2D (, ) λ 2 (+,+) λ 1 (+, ) 3D > <? λ1 λ3 λ3 λ3 λ1 λ3 λ1 λ3 λ3 λ3 λ1 λ1 λ1 λ3 > λ1 λ3 < λ1 > λ1 < Scaffold (prototype) 3D glyphs with 2D glyphs Find mapping from 3D tensor shapes to 3D superquadrics 4 $ %
12 4 $ %!i hides 4 discontinuity $ 2 1 slight shape problem 1 %
13 Meet the glyphs... (+,+,+) (+,+,-) (+,-,-) (-,-,-) Halos: improve visibility, colormap of tensor trace The Glyph Rendering Pipeline 4. Scale and rotate (modelview matrix) 1. Map! i to (u,v) 2. Map (u,v) to ($,%) 3. Generate Base Glyph 5. Color code sign (fragment shader)
14 Using a Palette of Base Geometry Precompute and re-use superquadric geometry on a grid in ($,%) space Implementation & Performance Geometry-based implementation Performance: 3 full-resolution glyphs 18x1 viewport 25fps (incl. halos) on NVIDIA Quadro FX 18 Most code is in Teem; please see tutorial on
15 Why not use Reynolds Glyphs? Colored Reynolds Glyphs clearly show negative eigenvalues unless you re viewing them from an unfortunate direction. Why Prefer Superquadrics? Superquadrics remain legible from a wide range of directions.
16 What About Traceless Tensors? Existing Traceless Superquadric Glyphs [Jankun-Kelly/Mehta, Vis6] Our new Superquadric Glyphs Hessians vs. Geometry Tensors Geometry Tensors Hessians See how geometry tensors are related to Hessians G (I-nn T )H(I-nn T )/ g
17 Inspecting Smoothing Image Filters Original Data Gaussian Smoothing Total Variation Flow Equal Remaining Variance What are Differences? Inspecting Smoothing Image Filters Original Data Gaussian Smoothing Total Variation Flow Let s look at Hessians
18 Inspecting Smoothing Image Filters Original Data Gaussian Smoothing Total Variation Flow Gaussian Filtering Smoothes Hessians. Leads to smearing out into flat regions and cancellation effects. Inspecting Smoothing Image Filters Original Data Gaussian Smoothing Total Variation Flow TV flow creates flat regions and sharp edges between them.
19 Results, turbulent jet flow LIC Results, turbulent jet flow LIC, modulate contrast by velocity
20 Results, turbulent jet flow Rate-of-Deformation Tensor ( v + T v)/2 Ellipses with! # exp(!) Results, turbulent jet flow Rate-of-Deformation Tensor ( v + T v)/2 New glyphs, s( D ) D
21 Results, turbulent jet flow Rate-of-Deformation Tensor ( v + T v)/2 New glyphs, s( D ) D 1/2 Results, double point stress field Can now see tensors underlying hyperstreamlines s( D ) D 1/1
22 Conclusions Presented a new set of tensor glyphs Can visualize all symmetric 3D 2nd-order tensors Glyphs design guided by principles Need disambiguity, avoid misleading symmetries Eigenplane projection invariance! scaffolds Future: still no satisfactory glyphs for: General 2nd-order (non-symmetric) tensors Rotations Acknowledgements Funding: DAAD Flow data: Wolfgang Kollmann, UC Davis Symmetry and Continuity discussion: 29 Dagstuhl Scientific Visualization Seminar 9251 Thank you
Medical Visualization - Tensor Visualization. J.-Prof. Dr. Kai Lawonn
Medical Visualization - Tensor Visualization J.-Prof. Dr. Kai Lawonn Lecture is partially based on the lecture by Prof. Thomas Schultz 2 What is a Tensor? A tensor is a multilinear transformation that
More informationSymmetry and Continuity in Visualization and Tensor Glyph Design. Gordon L. Kindlmann. Topic
Symmetry and Continuity in Visualization and Tensor Glyph Design Gordon L. Kindlmann Topic Symmetry and Continuity General: for colormaps, scalar vis... Specific: glyphs for symmetric tensors Experiences
More informationTensor Field Visualization. Ronald Peikert SciVis Tensor Fields 9-1
Tensor Field Visualization Ronald Peikert SciVis 2007 - Tensor Fields 9-1 Tensors "Tensors are the language of mechanics" Tensor of order (rank) 0: scalar 1: vector 2: matrix (example: stress tensor) Tensors
More informationTensor fields. Tensor fields: Outline. Chantal Oberson Ausoni
Tensor fields Chantal Oberson Ausoni 7.8.2014 ICS Summer school Roscoff - Visualization at the interfaces 28.7-8.8, 2014 1 Tensor fields: Outline 1. TENSOR FIELDS: DEFINITION 2. PROPERTIES OF SECOND-ORDER
More informationMaking sense of Math in Vis. Gordon Kindlmann University of Chicago
Making sense of Math in Vis Gordon Kindlmann University of Chicago glk@uchicago.edu (from seminar description) http://www.dagstuhl.de/en/program/calendar/semhp/?semnr=18041 Mathematical foundations of
More informationTensor Visualization. CSC 7443: Scientific Information Visualization
Tensor Visualization Tensor data A tensor is a multivariate quantity Scalar is a tensor of rank zero s = s(x,y,z) Vector is a tensor of rank one v = (v x,v y,v z ) For a symmetric tensor of rank 2, its
More information1 Diffusion Tensor. x 1, , x n
Tensor Field Visualization Tensor is the extension of concept of scalar and vector, it is the language of mechanics. Therefore, tensor field visualization is a challenging issue for scientific visualization.
More informationTensor Visualisation
Tensor Visualisation Computer Animation and Visualisation Lecture 18 tkomura@ed.ac.uk Institute for Perception, Action & Behaviour School of Informatics Tensors 1 Reminder : Attribute Data Types Scalar
More informationTensor Visualisation
Tensor Visualisation Computer Animation and Visualisation Lecture 16 Taku Komura tkomura@ed.ac.uk Institute for Perception, Action & Behaviour School of Informatics 1 Tensor Visualisation What is tensor
More informationNONLINEAR DIFFUSION PDES
NONLINEAR DIFFUSION PDES Erkut Erdem Hacettepe University March 5 th, 0 CONTENTS Perona-Malik Type Nonlinear Diffusion Edge Enhancing Diffusion 5 References 7 PERONA-MALIK TYPE NONLINEAR DIFFUSION The
More informationGlyph-based Comparative Visualization for Diffusion Tensor Fields
797 IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS, VOL. 22, NO. 1, JANUARY 2016 Glyph-based Comparative Visualization for Diffusion Tensor Fields Changgong Zhang, Thomas Schultz, Kai Lawonn,
More informationA Visual Approach to Analysis of Stress Tensor Fields
A Visual Approach to Analysis of Stress Tensor Fields Andrea Kratz 1, Björn Meyer 1, and Ingrid Hotz 1 1 Zuse Institute Berlin (ZIB) Department Visualization and Data Analysis {kratz, bjoern.meyer, hotz}@zib.de
More informationScientific Visualization at University of Chicago. Gordon L. Kindlmann Let!s recap some Calculus Taylor expansion of scalar field
Scientific Visualization at University of Chicago Gordon L. Kindlmann Let!s recap some Calculus Taylor expansion of scalar field f(x 0 + )=f(x 0 )+ f(x 0 ) + o( ) x 0 f(x 0 ) f(x 0 )
More informationLecture 11: Differential Geometry
Lecture 11: Differential Geometry c Bryan S. Morse, Brigham Young University, 1998 2000 Last modified on February 28, 2000 at 8:45 PM Contents 11.1 Introduction..............................................
More information, b = 0. (2) 1 2 The eigenvectors of A corresponding to the eigenvalues λ 1 = 1, λ 2 = 3 are
Quadratic forms We consider the quadratic function f : R 2 R defined by f(x) = 2 xt Ax b T x with x = (x, x 2 ) T, () where A R 2 2 is symmetric and b R 2. We will see that, depending on the eigenvalues
More informationG12MAN Mathematical Analysis Fully-justified answers to Revision Quiz questions
G12MAN Mathematical Analysis Fully-justified answers to Revision Quiz questions Remember that, unless otherwise specified or not applicable for some reason), you MUST justify your answers to questions
More informationMath (P)Review Part II:
Math (P)Review Part II: Vector Calculus Computer Graphics Assignment 0.5 (Out today!) Same story as last homework; second part on vector calculus. Slightly fewer questions Last Time: Linear Algebra Touched
More information2D Asymmetric Tensor Analysis
2D Asymmetric Tensor Analysis Xiaoqiang Zheng Alex Pang Computer Science Department University of California, Santa Cruz, CA 95064 ABSTRACT Analysis of degenerate tensors is a fundamental step in finding
More informationTensor Visualisation
Tensor Visualisation Computer Animation and Visualisation Lecture 15 Taku Komura tkomura@ed.ac.uk Institute for Perception, Action & Behaviour School of Informatics 1 Overview Tensor Visualisation What
More informationStress, Strain, Mohr s Circle
Stress, Strain, Mohr s Circle The fundamental quantities in solid mechanics are stresses and strains. In accordance with the continuum mechanics assumption, the molecular structure of materials is neglected
More informationFluid Dynamics Exercises and questions for the course
Fluid Dynamics Exercises and questions for the course January 15, 2014 A two dimensional flow field characterised by the following velocity components in polar coordinates is called a free vortex: u r
More informationLecture 8: Interest Point Detection. Saad J Bedros
#1 Lecture 8: Interest Point Detection Saad J Bedros sbedros@umn.edu Last Lecture : Edge Detection Preprocessing of image is desired to eliminate or at least minimize noise effects There is always tradeoff
More informationAPPENDIX A. Background Mathematics. A.1 Linear Algebra. Vector algebra. Let x denote the n-dimensional column vector with components x 1 x 2.
APPENDIX A Background Mathematics A. Linear Algebra A.. Vector algebra Let x denote the n-dimensional column vector with components 0 x x 2 B C @. A x n Definition 6 (scalar product). The scalar product
More informationHW1 solutions. 1. α Ef(x) β, where Ef(x) is the expected value of f(x), i.e., Ef(x) = n. i=1 p if(a i ). (The function f : R R is given.
HW1 solutions Exercise 1 (Some sets of probability distributions.) Let x be a real-valued random variable with Prob(x = a i ) = p i, i = 1,..., n, where a 1 < a 2 < < a n. Of course p R n lies in the standard
More informationTurbulence Modeling I!
Outline! Turbulence Modeling I! Grétar Tryggvason! Spring 2010! Why turbulence modeling! Reynolds Averaged Numerical Simulations! Zero and One equation models! Two equations models! Model predictions!
More informationCorner detection: the basic idea
Corner detection: the basic idea At a corner, shifting a window in any direction should give a large change in intensity flat region: no change in all directions edge : no change along the edge direction
More informationTensor Visualisation and Information Visualisation
Tensor Visualisation and Information Visualisation Computer Animation and Visualisation Lecture 18 Taku Komura tkomura@ed.ac.uk Institute for Perception, Action & Behaviour School of Informatics 1 Reminder
More informationLecture 8: Interest Point Detection. Saad J Bedros
#1 Lecture 8: Interest Point Detection Saad J Bedros sbedros@umn.edu Review of Edge Detectors #2 Today s Lecture Interest Points Detection What do we mean with Interest Point Detection in an Image Goal:
More informationChapter 9: Differential Analysis
9-1 Introduction 9-2 Conservation of Mass 9-3 The Stream Function 9-4 Conservation of Linear Momentum 9-5 Navier Stokes Equation 9-6 Differential Analysis Problems Recall 9-1 Introduction (1) Chap 5: Control
More informationThis paper is about a new method for analyzing the distribution property of nonuniform stochastic samples.
This paper is about a new method for analyzing the distribution property of nonuniform stochastic samples. Stochastic sampling is a fundamental component in many graphics applications. Examples include
More informationEdges and Scale. Image Features. Detecting edges. Origin of Edges. Solution: smooth first. Effects of noise
Edges and Scale Image Features From Sandlot Science Slides revised from S. Seitz, R. Szeliski, S. Lazebnik, etc. Origin of Edges surface normal discontinuity depth discontinuity surface color discontinuity
More informationAnalysis of Functions
Lecture for Week 11 (Secs. 5.1 3) Analysis of Functions (We used to call this topic curve sketching, before students could sketch curves by typing formulas into their calculators. It is still important
More information(Refer Slide Time: 2:14)
Fluid Dynamics And Turbo Machines. Professor Dr Shamit Bakshi. Department Of Mechanical Engineering. Indian Institute Of Technology Madras. Part A. Module-1. Lecture-3. Introduction To Fluid Flow. (Refer
More informationScaling the Topology of Symmetric, Second-Order Planar Tensor Fields
Scaling the Topology of Symmetric, Second-Order Planar Tensor Fields Xavier Tricoche, Gerik Scheuermann, and Hans Hagen University of Kaiserslautern, P.O. Box 3049, 67653 Kaiserslautern, Germany E-mail:
More informationChapter 9: Differential Analysis of Fluid Flow
of Fluid Flow Objectives 1. Understand how the differential equations of mass and momentum conservation are derived. 2. Calculate the stream function and pressure field, and plot streamlines for a known
More informationLecture 3: The Navier-Stokes Equations: Topological aspects
Lecture 3: The Navier-Stokes Equations: Topological aspects September 9, 2015 1 Goal Topology is the branch of math wich studies shape-changing objects; objects which can transform one into another without
More informationMath (P)refresher Lecture 8: Unconstrained Optimization
Math (P)refresher Lecture 8: Unconstrained Optimization September 2006 Today s Topics : Quadratic Forms Definiteness of Quadratic Forms Maxima and Minima in R n First Order Conditions Second Order Conditions
More informationImplicit sampling for particle filters. Alexandre Chorin, Mathias Morzfeld, Xuemin Tu, Ethan Atkins
0/20 Implicit sampling for particle filters Alexandre Chorin, Mathias Morzfeld, Xuemin Tu, Ethan Atkins University of California at Berkeley 2/20 Example: Try to find people in a boat in the middle of
More informationMASSACHUSETTS INSTITUTE OF TECHNOLOGY DEPARTMENT OF MATERIALS SCIENCE AND ENGINEERING CAMBRIDGE, MASSACHUSETTS 02139
MASSACHUSETTS NSTTUTE OF TECHNOLOGY DEPARTMENT OF MATERALS SCENCE AND ENGNEERNG CAMBRDGE, MASSACHUSETTS 39 3. MECHANCAL PROPERTES OF MATERALS PROBLEM SET SOLUTONS Reading Ashby, M.F., 98, Tensors: Notes
More informationExtract useful building blocks: blobs. the same image like for the corners
Extract useful building blocks: blobs the same image like for the corners Here were the corners... Blob detection in 2D Laplacian of Gaussian: Circularly symmetric operator for blob detection in 2D 2 g=
More informationRANDOM PROPERTIES BENOIT PAUSADER
RANDOM PROPERTIES BENOIT PAUSADER. Quasilinear problems In general, one consider the following trichotomy for nonlinear PDEs: A semilinear problem is a problem where the highest-order terms appears linearly
More informationRelaxation methods and finite element schemes for the equations of visco-elastodynamics. Chiara Simeoni
Relaxation methods and finite element schemes for the equations of visco-elastodynamics Chiara Simeoni Department of Information Engineering, Computer Science and Mathematics University of L Aquila (Italy)
More informationTable of Contents. II. PDE classification II.1. Motivation and Examples. II.2. Classification. II.3. Well-posedness according to Hadamard
Table of Contents II. PDE classification II.. Motivation and Examples II.2. Classification II.3. Well-posedness according to Hadamard Chapter II (ContentChapterII) Crashtest: Reality Simulation http:www.ara.comprojectssvocrownvic.htm
More informationUNIVERSITY of LIMERICK OLLSCOIL LUIMNIGH
UNIVERSITY of LIMERICK OLLSCOIL LUIMNIGH College of Informatics and Electronics END OF SEMESTER ASSESSMENT PAPER MODULE CODE: MS4613 SEMESTER: Autumn 2002/03 MODULE TITLE: Vector Analysis DURATION OF EXAMINATION:
More information(Refer Slide Time: 0:35)
Fluid Dynamics And Turbo Machines. Professor Dr Shamit Bakshi. Department Of Mechanical Engineering. Indian Institute Of Technology Madras. Part A. Module-1. Lecture-4. Tutorial. (Refer Slide Time: 0:35)
More informationN. L. P. NONLINEAR PROGRAMMING (NLP) deals with optimization models with at least one nonlinear function. NLP. Optimization. Models of following form:
0.1 N. L. P. Katta G. Murty, IOE 611 Lecture slides Introductory Lecture NONLINEAR PROGRAMMING (NLP) deals with optimization models with at least one nonlinear function. NLP does not include everything
More informationCS 4204 Computer Graphics
CS 4204 Computer Graphics Vector and Matrix Yong Cao Virginia Tech Vectors N-tuple: Vectors N-tuple tuple: Magnitude: Unit vectors Normalizing a vector Operations with vectors Addition Multiplication with
More informationMath RE - Calculus II Antiderivatives and the Indefinite Integral Page 1 of 5
Math 201-203-RE - Calculus II Antiderivatives and the Indefinite Integral Page 1 of 5 What is the Antiderivative? In a derivative problem, a function f(x) is given and you find the derivative f (x) using
More informationCurve Fitting. 1 Interpolation. 2 Composite Fitting. 1.1 Fitting f(x) 1.2 Hermite interpolation. 2.1 Parabolic and Cubic Splines
Curve Fitting Why do we want to curve fit? In general, we fit data points to produce a smooth representation of the system whose response generated the data points We do this for a variety of reasons 1
More informationThe GET Operator. LiTH-ISY-R-2633
The GET Operator Michael Felsberg Linköping University, Computer Vision Laboratory, SE-58183 Linköping, Sweden, mfe@isy.liu.se, http://www.isy.liu.se/cvl/ October 4, 004 LiTH-ISY-R-633 Abstract In this
More informationLine of symmetry Total area enclosed is 1
The Normal distribution The Normal distribution is a continuous theoretical probability distribution and, probably, the most important distribution in Statistics. Its name is justified by the fact that
More informationShape of Gaussians as Feature Descriptors
Shape of Gaussians as Feature Descriptors Liyu Gong, Tianjiang Wang and Fang Liu Intelligent and Distributed Computing Lab, School of Computer Science and Technology Huazhong University of Science and
More informationHW3 - Due 02/06. Each answer must be mathematically justified. Don t forget your name. 1 2, A = 2 2
HW3 - Due 02/06 Each answer must be mathematically justified Don t forget your name Problem 1 Find a 2 2 matrix B such that B 3 = A, where A = 2 2 If A was diagonal, it would be easy: we would just take
More informationVisualizing Uncertainty in HARDI Tractography Using Superquadric Streamtubes
Eurographics Conference on Visualization (EuroVis) (2014) N. Elmqvist, M. Hlawitschka, and J. Kennedy (Editors) Short Papers Visualizing Uncertainty in HARDI Tractography Using Superquadric Streamtubes
More informationEstimators for Orientation and Anisotropy in Digitized Images
Estimators for Orientation and Anisotropy in Digitized Images Lucas J. van Vliet and Piet W. Verbeek Pattern Recognition Group of the Faculty of Applied Physics Delft University of Technolo Lorentzweg,
More informationEffects of large-scale free stream turbulence on a turbulent boundary layer
Effects of large-scale free stream turbulence on a turbulent boundary layer Nicole S. Sharp Cornell University Presented in partial completion of the requirements for the degree of Master of Science in
More informationPHYM432 Relativity and Cosmology 17. Cosmology Robertson Walker Metric
PHYM432 Relativity and Cosmology 17. Cosmology Robertson Walker Metric Cosmology applies physics to the universe as a whole, describing it s origin, nature evolution and ultimate fate. While these questions
More informationFeature extraction: Corners and blobs
Feature extraction: Corners and blobs Review: Linear filtering and edge detection Name two different kinds of image noise Name a non-linear smoothing filter What advantages does median filtering have over
More informationRecap: edge detection. Source: D. Lowe, L. Fei-Fei
Recap: edge detection Source: D. Lowe, L. Fei-Fei Canny edge detector 1. Filter image with x, y derivatives of Gaussian 2. Find magnitude and orientation of gradient 3. Non-maximum suppression: Thin multi-pixel
More informationQuantitative Techniques (Finance) 203. Polynomial Functions
Quantitative Techniques (Finance) 03 Polynomial Functions Felix Chan October 006 Introduction This topic discusses the properties and the applications of polynomial functions, specifically, linear and
More informationLES of turbulent shear flow and pressure driven flow on shallow continental shelves.
LES of turbulent shear flow and pressure driven flow on shallow continental shelves. Guillaume Martinat,CCPO - Old Dominion University Chester Grosch, CCPO - Old Dominion University Ying Xu, Michigan State
More informationMultidimensional Geometry and its Applications
PARALLEL COORDINATES : VISUAL Multidimensional Geometry and its Applications Alfred Inselberg( c 99, ) Senior Fellow San Diego SuperComputing Center, CA, USA Computer Science and Applied Mathematics Departments
More informationPerspectives on Reynolds Stress Modeling Part I: General Approach. Umich/NASA Symposium on Advances in Turbulence Modeling
DLR.de Chart 1 Perspectives on Reynolds Stress Modeling Part I: General Approach Umich/NASA Symposium on Advances in Turbulence Modeling Bernhard Eisfeld, Tobias Knopp 11 July 2017 DLR.de Chart 2 Overview
More informationGeometric Modeling Summer Semester 2012 Linear Algebra & Function Spaces
Geometric Modeling Summer Semester 2012 Linear Algebra & Function Spaces (Recap) Announcement Room change: On Thursday, April 26th, room 024 is occupied. The lecture will be moved to room 021, E1 4 (the
More informationFFTs in Graphics and Vision. Groups and Representations
FFTs in Graphics and Vision Groups and Representations Outline Groups Representations Schur s Lemma Correlation Groups A group is a set of elements G with a binary operation (often denoted ) such that
More informationSoft Bodies. Good approximation for hard ones. approximation breaks when objects break, or deform. Generalization: soft (deformable) bodies
Soft-Body Physics Soft Bodies Realistic objects are not purely rigid. Good approximation for hard ones. approximation breaks when objects break, or deform. Generalization: soft (deformable) bodies Deformed
More informationGradient Methods Using Momentum and Memory
Chapter 3 Gradient Methods Using Momentum and Memory The steepest descent method described in Chapter always steps in the negative gradient direction, which is orthogonal to the boundary of the level set
More informationDesigning Information Devices and Systems I Spring 2016 Official Lecture Notes Note 21
EECS 6A Designing Information Devices and Systems I Spring 26 Official Lecture Notes Note 2 Introduction In this lecture note, we will introduce the last topics of this semester, change of basis and diagonalization.
More informationNon-convex optimization. Issam Laradji
Non-convex optimization Issam Laradji Strongly Convex Objective function f(x) x Strongly Convex Objective function Assumptions Gradient Lipschitz continuous f(x) Strongly convex x Strongly Convex Objective
More informationMachine vision. Summary # 4. The mask for Laplacian is given
1 Machine vision Summary # 4 The mask for Laplacian is given L = 0 1 0 1 4 1 (6) 0 1 0 Another Laplacian mask that gives more importance to the center element is L = 1 1 1 1 8 1 (7) 1 1 1 Note that the
More informationDependence. MFM Practitioner Module: Risk & Asset Allocation. John Dodson. September 11, Dependence. John Dodson. Outline.
MFM Practitioner Module: Risk & Asset Allocation September 11, 2013 Before we define dependence, it is useful to define Random variables X and Y are independent iff For all x, y. In particular, F (X,Y
More informationArtificial Intelligence Module 2. Feature Selection. Andrea Torsello
Artificial Intelligence Module 2 Feature Selection Andrea Torsello We have seen that high dimensional data is hard to classify (curse of dimensionality) Often however, the data does not fill all the space
More informationComparison of Models for Finite Plasticity
Comparison of Models for Finite Plasticity A numerical study Patrizio Neff and Christian Wieners California Institute of Technology (Universität Darmstadt) Universität Augsburg (Universität Heidelberg)
More informationInteractive Visual Analysis of Multi-faceted Scientific Data Johannes Kehrer
Interactive Visual Analysis of Multi-faceted Scientific Data Johannes Kehrer Visualization Group, Dept. of Informatics University of Bergen, Norway www.ii.uib.no/vis Motivation Increasing amounts of scientific
More informationWed. Sept 28th: 1.3 New Functions from Old Functions: o vertical and horizontal shifts o vertical and horizontal stretching and reflecting o
Homework: Appendix A: 1, 2, 3, 5, 6, 7, 8, 11, 13-33(odd), 34, 37, 38, 44, 45, 49, 51, 56. Appendix B: 3, 6, 7, 9, 11, 14, 16-21, 24, 29, 33, 36, 37, 42. Appendix D: 1, 2, 4, 9, 11-20, 23, 26, 28, 29,
More informationTensor Field Visualization Using a Metric Interpretation
Tensor Field Visualization Using a Metric Interpretation Ingrid Hotz, Louis Feng, Hans Hagen 2, Bernd Hamann, and Kenneth Joy Institute for Data Analysis and Visualization, (IDAV), Department of Computer
More informationA PRIMER ON SESQUILINEAR FORMS
A PRIMER ON SESQUILINEAR FORMS BRIAN OSSERMAN This is an alternative presentation of most of the material from 8., 8.2, 8.3, 8.4, 8.5 and 8.8 of Artin s book. Any terminology (such as sesquilinear form
More informationA Dielectric Invisibility Carpet
A Dielectric Invisibility Carpet Jensen Li Prof. Xiang Zhang s Research Group Nanoscale Science and Engineering Center (NSEC) University of California at Berkeley, USA CLK08-09/22/2008 Presented at Center
More informationParameter estimation Conditional risk
Parameter estimation Conditional risk Formalizing the problem Specify random variables we care about e.g., Commute Time e.g., Heights of buildings in a city We might then pick a particular distribution
More information1) the intermittence of the vortex-shedding regime at the critical angle of incidence in smooth flow; ) the inversion of the lift coefficient slope at
The Seventh International Colloquium on Bluff Body Aerodynamics and Applications (BBAA7) Shanghai, China; September -6, 01 Experimental investigation on the aerodynamic behavior of square cylinders with
More informationFrequency, Vibration, and Fourier
Lecture 22: Frequency, Vibration, and Fourier Computer Graphics CMU 15-462/15-662, Fall 2015 Last time: Numerical Linear Algebra Graphics via linear systems of equations Why linear? Have to solve BIG problems
More informationGaussian processes. Chuong B. Do (updated by Honglak Lee) November 22, 2008
Gaussian processes Chuong B Do (updated by Honglak Lee) November 22, 2008 Many of the classical machine learning algorithms that we talked about during the first half of this course fit the following pattern:
More informationLecture 6 Positive Definite Matrices
Linear Algebra Lecture 6 Positive Definite Matrices Prof. Chun-Hung Liu Dept. of Electrical and Computer Engineering National Chiao Tung University Spring 2017 2017/6/8 Lecture 6: Positive Definite Matrices
More informationTime scales. Notes. Mixed Implicit/Explicit. Newmark Methods [ ] [ ] "!t = O 1 %
Notes Time scales! For using Pixie (the renderer) make sure you type use pixie first! Assignment 1 questions?! [work out]! For position dependence, characteristic time interval is "!t = O 1 % $ ' # K &!
More informationEugene Zhang Oregon State University
2D Asymmetric Tensor Field Analysis and Visualization Eugene Zhang Oregon State University Introduction Asymmetric tensors can model the gradient of a vector field velocity gradient in fluid dynamics deformation
More informationLagrange Multipliers
Optimization with Constraints As long as algebra and geometry have been separated, their progress have been slow and their uses limited; but when these two sciences have been united, they have lent each
More informationOR MSc Maths Revision Course
OR MSc Maths Revision Course Tom Byrne School of Mathematics University of Edinburgh t.m.byrne@sms.ed.ac.uk 15 September 2017 General Information Today JCMB Lecture Theatre A, 09:30-12:30 Mathematics revision
More informationImage enhancement. Why image enhancement? Why image enhancement? Why image enhancement? Example of artifacts caused by image encoding
13 Why image enhancement? Image enhancement Example of artifacts caused by image encoding Computer Vision, Lecture 14 Michael Felsberg Computer Vision Laboratory Department of Electrical Engineering 12
More informationCFD in Heat Transfer Equipment Professor Bengt Sunden Division of Heat Transfer Department of Energy Sciences Lund University
CFD in Heat Transfer Equipment Professor Bengt Sunden Division of Heat Transfer Department of Energy Sciences Lund University email: bengt.sunden@energy.lth.se CFD? CFD = Computational Fluid Dynamics;
More informationIntroducing the Laplacian
Introducing the Laplacian Prepared by: Steven Butler October 3, 005 A look ahead to weighted edges To this point we have looked at random walks where at each vertex we have equal probability of moving
More informationCalculus Honors Curriculum Guide Dunmore School District Dunmore, PA
Calculus Honors Dunmore School District Dunmore, PA Calculus Honors Prerequisite: Successful completion of Trigonometry/Pre-Calculus Honors Major topics include: limits, derivatives, integrals. Instruction
More informationMachine vision, spring 2018 Summary 4
Machine vision Summary # 4 The mask for Laplacian is given L = 4 (6) Another Laplacian mask that gives more importance to the center element is given by L = 8 (7) Note that the sum of the elements in the
More informationNeural Network Training
Neural Network Training Sargur Srihari Topics in Network Training 0. Neural network parameters Probabilistic problem formulation Specifying the activation and error functions for Regression Binary classification
More informationDefinition and basic properties of heat kernels I, An introduction
Definition and basic properties of heat kernels I, An introduction Zhiqin Lu, Department of Mathematics, UC Irvine, Irvine CA 92697 April 23, 2010 In this lecture, we will answer the following questions:
More informationFunction Approximation
1 Function Approximation This is page i Printer: Opaque this 1.1 Introduction In this chapter we discuss approximating functional forms. Both in econometric and in numerical problems, the need for an approximating
More informationLECTURE NOTE #11 PROF. ALAN YUILLE
LECTURE NOTE #11 PROF. ALAN YUILLE 1. NonLinear Dimension Reduction Spectral Methods. The basic idea is to assume that the data lies on a manifold/surface in D-dimensional space, see figure (1) Perform
More informationTHE DOUBLE COVER RELATIVE TO A CONVEX DOMAIN AND THE RELATIVE ISOPERIMETRIC INEQUALITY
J. Aust. Math. Soc. 80 (2006), 375 382 THE DOUBLE COVER RELATIVE TO A CONVEX DOMAIN AND THE RELATIVE ISOPERIMETRIC INEQUALITY JAIGYOUNG CHOE (Received 18 March 2004; revised 16 February 2005) Communicated
More information3. Numerical integration
3. Numerical integration... 3. One-dimensional quadratures... 3. Two- and three-dimensional quadratures... 3.3 Exact Integrals for Straight Sided Triangles... 5 3.4 Reduced and Selected Integration...
More informationWavelet-based Salient Points with Scale Information for Classification
Wavelet-based Salient Points with Scale Information for Classification Alexandra Teynor and Hans Burkhardt Department of Computer Science, Albert-Ludwigs-Universität Freiburg, Germany {teynor, Hans.Burkhardt}@informatik.uni-freiburg.de
More informationFeature Vector Similarity Based on Local Structure
Feature Vector Similarity Based on Local Structure Evgeniya Balmachnova, Luc Florack, and Bart ter Haar Romeny Eindhoven University of Technology, P.O. Box 53, 5600 MB Eindhoven, The Netherlands {E.Balmachnova,L.M.J.Florack,B.M.terHaarRomeny}@tue.nl
More information