Parameter estimation class 5

Similar documents
Computer Vision. The 2D projective plane and it s applications. HZ Ch 2. In particular: Ch 2.1-4, 2.7, Szelisky: Ch 2.1.1, 2.1.2

CS 532: 3D Computer Vision 2 nd Set of Notes

Structure from Motion. Forsyth&Ponce: Chap. 12 and 13 Szeliski: Chap. 7

Linear discriminants. Nuno Vasconcelos ECE Department, UCSD

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CS4495/6495 Introduction to Computer Vision. 3C-L3 Calibrating cameras

Minimizing Algebraic Error in Geometric Estimation Problems

MACHINE APPLIED MACHINE LEARNING LEARNING. Gaussian Mixture Regression

e i is a random error

Pattern Classification

where λ = Z/f. where a 3 4 projection matrix represents a map from 3D to 2D. Part I: Single and Two View Geometry Internal camera parameters

Image classification. Given the bag-of-features representations of images from different classes, how do we learn a model for distinguishing i them?

Which Separator? Spring 1

CIS526: Machine Learning Lecture 3 (Sept 16, 2003) Linear Regression. Preparation help: Xiaoying Huang. x 1 θ 1 output... θ M x M

Chapter 2 - The Simple Linear Regression Model S =0. e i is a random error. S β2 β. This is a minimization problem. Solution is a calculus exercise.

A Tutorial on Data Reduction. Linear Discriminant Analysis (LDA) Shireen Elhabian and Aly A. Farag. University of Louisville, CVIP Lab September 2009

Estimating the Fundamental Matrix by Transforming Image Points in Projective Space 1

Discriminative classifier: Logistic Regression. CS534-Machine Learning

OPTIMISATION. Introduction Single Variable Unconstrained Optimisation Multivariable Unconstrained Optimisation Linear Programming

Linear Regression Introduction to Machine Learning. Matt Gormley Lecture 5 September 14, Readings: Bishop, 3.1

15-381: Artificial Intelligence. Regression and cross validation

Report on Image warping

( ) [ ( k) ( k) ( x) ( ) ( ) ( ) [ ] ξ [ ] [ ] [ ] ( )( ) i ( ) ( )( ) 2! ( ) = ( ) 3 Interpolation. Polynomial Approximation.

Shuai Dong. Isaac Newton. Gottfried Leibniz

Hierarchical State Estimation Using Phasor Measurement Units

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Errors for Linear Systems

n α j x j = 0 j=1 has a nontrivial solution. Here A is the n k matrix whose jth column is the vector for all t j=0

Tracking with Kalman Filter

Linear Approximation with Regularization and Moving Least Squares

Lecture 16 Statistical Analysis in Biomaterials Research (Part II)

Discriminative classifier: Logistic Regression. CS534-Machine Learning

Machine Learning for Signal Processing Linear Gaussian Models

Kernel Methods and SVMs Extension

VQ widely used in coding speech, image, and video

COMPLEX NUMBERS AND QUADRATIC EQUATIONS

Lecture Notes on Linear Regression

ADVANCED MACHINE LEARNING ADVANCED MACHINE LEARNING

1 Matrix representations of canonical matrices

Machine Learning for Signal Processing Linear Gaussian Models

Lecture 10 Support Vector Machines II

Not-for-Publication Appendix to Optimal Asymptotic Least Aquares Estimation in a Singular Set-up

Department of Quantitative Methods & Information Systems. Time Series and Their Components QMIS 320. Chapter 6

Polynomial Regression Models

Chapter 9: Statistical Inference and the Relationship between Two Variables

Instituto Tecnológico de Aeronáutica FINITE ELEMENTS I. Class notes AE-245

Maximum Likelihood Estimation of Binary Dependent Variables Models: Probit and Logit. 1. General Formulation of Binary Dependent Variables Models

Generalized Linear Methods

Review: Fit a line to N data points

Correlation and Regression. Correlation 9.1. Correlation. Chapter 9

Systems of Equations (SUR, GMM, and 3SLS)

Stanford University CS254: Computational Complexity Notes 7 Luca Trevisan January 29, Notes for Lecture 7

The Karush-Kuhn-Tucker. Nuno Vasconcelos ECE Department, UCSD

Lecture 10: Dimensionality reduction

17. Coordinate-Free Projective Geometry for Computer Vision

Inner Product. Euclidean Space. Orthonormal Basis. Orthogonal

CS 3710: Visual Recognition Classification and Detection. Adriana Kovashka Department of Computer Science January 13, 2015

Statistics for Economics & Business

Lecture Nov

Solution for singularly perturbed problems via cubic spline in tension

Maximum Likelihood Estimation (MLE)

Support Vector Machines

The Gaussian classifier. Nuno Vasconcelos ECE Department, UCSD

Probability Theory. The nth coefficient of the Taylor series of f(k), expanded around k = 0, gives the nth moment of x as ( ik) n n!

Support Vector Machines. Vibhav Gogate The University of Texas at dallas

Rockefeller College University at Albany

PART 8. Partial Differential Equations PDEs

Solutions to exam in SF1811 Optimization, Jan 14, 2015

Multivariate Ratio Estimator of the Population Total under Stratified Random Sampling

Multigrid Methods and Applications in CFD

Finite Difference Method

Support Vector Machines

LINEAR REGRESSION ANALYSIS. MODULE IX Lecture Multicollinearity

Meshless Surfaces. presented by Niloy J. Mitra. An Nguyen

Probability-Theoretic Junction Trees

Systematic Error Illustration of Bias. Sources of Systematic Errors. Effects of Systematic Errors 9/23/2009. Instrument Errors Method Errors Personal

CS 523: Computer Graphics, Spring Shape Modeling. PCA Applications + SVD. Andrew Nealen, Rutgers, /15/2011 1

1 GSW Iterative Techniques for y = Ax

Joint Statistical Meetings - Biopharmaceutical Section

4.3 Poisson Regression

15.1 The geometric basis for the trifocal tensor Incidence relations for lines.

A New Recursive Method for Solving State Equations Using Taylor Series

Multiple-Linear, Polynomial and Nonlinear Regression Basic Concepts (1)

Grid Generation around a Cylinder by Complex Potential Functions

3.1 ML and Empirical Distribution

Maximal Margin Classifier

Thermal tomography on the basis of an information method

Hydrological statistics. Hydrological statistics and extremes

Some Comments on Accelerating Convergence of Iterative Sequences Using Direct Inversion of the Iterative Subspace (DIIS)

Norms, Condition Numbers, Eigenvalues and Eigenvectors

Resource Allocation and Decision Analysis (ECON 8010) Spring 2014 Foundations of Regression Analysis

36.1 Why is it important to be able to find roots to systems of equations? Up to this point, we have discussed how to find the solution to

Chapter 8 Indicator Variables

The Fundamental Theorem of Algebra. Objective To use the Fundamental Theorem of Algebra to solve polynomial equations with complex solutions

Lecture 12: Classification

15 Lagrange Multipliers

Why BP Works STAT 232B

Classification as a Regression Problem

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Lecture 6/7 (February 10/12, 2014) DIRAC EQUATION. The non-relativistic Schrödinger equation was obtained by noting that the Hamiltonian 2

Transcription:

Parameter estmaton class 5 Multple Ve Geometr Comp 9-89 Marc Pollefes

Content Background: Projectve geometr (D, 3D), Parameter estmaton, Algortm evaluaton. Sngle Ve: Camera model, Calbraton, Sngle Ve Geometr. o Ves: Eppolar Geometr, 3D reconstructon, Computng F, Computng structure, Plane and omograpes. ree Ves: rfocal ensor, Computng. More Ves: N-Lneartes, Multple ve reconstructon, Bundle adjustment, autocalbraton, Dnamc SfM, Ceralt, Dualt

Multple Ve Geometr course scedule (subject to cange) Jan. 7, 9 Intro & motvaton Projectve D Geometr Jan. 4, 6 (no class) Projectve D Geometr Jan., 3 Projectve 3D Geometr (no class) Jan. 8, 3 Parameter Estmaton Parameter Estmaton Feb. 4, 6 Algortm Evaluaton Camera Models Feb., 3 Camera Calbraton Sngle Ve Geometr Feb. 8, Eppolar Geometr 3D reconstructon Feb. 5, 7 Fund. Matr Comp. Structure Comp. Mar. 4, 6 Planes & Homograpes rfocal ensor Mar. 8, ree Ve Reconstructon Multple Ve Geometr Mar. 5, 7 MultpleVe Reconstructon Bundle adjustment Apr., 3 Auto-Calbraton Papers Apr. 8, Dnamc SfM Papers Apr. 5, 7 Ceralt Papers Apr., 4 Dualt Project Demos

Projectve 3D Geometr Ponts, lnes, planes and quadrcs ransformatons П, ω and Ω

Sngular Value Decomposton A UΣ V UΣΣ V Homogeneous least-squares mn A subject to soluton Vn

Parameter estmaton D omograp Gven a set of (, ), compute H ( H ) 3D to D camera projecton Gven a set of (, ), compute P ( P ) Fundamental matr Gven a set of (, ), compute F ( F ) rfocal tensor Gven a set of (,, ), compute

Number of measurements requred At least as man ndependent equatons as degrees of freedom requred Eample: λ ' 3 H 3 3 3 33 ndependent equatons / pont 8 degrees of freedom 4 8

Appromate solutons Mnmal soluton 4 ponts eld an eact soluton for H More ponts No eact soluton, because measurements are neact ( nose ) Searc for best accordng to some cost functon Algebrac or geometrc/statstcal cost

Gold Standard algortm Cost functon tat s optmal for some assumptons Computatonal algortm tat mnmzes t s called Gold Standard algortm Oter algortms can ten be compared to t

Drect Lnear ransformaton (DL) H H H 3 H 3 3 3 ( ),, A

Drect Lnear ransformaton (DL) Equatons are lnear n 3 A A A 3 + + A Onl out of 3 are lnearl ndependent (ndeed, eq/pt) 3 (onl drop trd ro f ) Holds for an omogeneous representaton, e.g. (,,)

Drect Lnear ransformaton Solvng for H (DL) A A A A 3 A4 sze A s 89 or 9, but rank 8 rval soluton s 9 s not nterestng -D null-space elds soluton of nterest pck for eample te one t

Drect Lnear ransformaton (DL) Over-determned soluton No eact soluton because of neact measurement.e. nose A A A M An Fnd appromate soluton - Addtonal constrant needed to avod, e.g. A - not possble, so mnmze A

DL algortm Objectve Gven n 4 D to D pont correspondences { }, determne te D omograp matr H suc tat H Algortm () () For eac correspondence compute A. Usuall onl to frst ros needed. Assemble n 9 matrces A nto a sngle n9 matr A () Obtan SVD of A. Soluton for s last column of V (v) Determne H from

Inomogeneous soluton ' ' ~ ' ' ' ' ' ' ' ' ' ' Snce can onl be computed up to scale, pck j, e.g. 9, and solve for 8-vector ~ Solve usng Gaussan elmnaton (4 ponts) or usng lnear least-squares (more tan 4 ponts) Hoever, f 9 ts approac fals also poor results f 9 close to zero erefore, not recommended Note 9 H 33 f orgn s mapped to nfnt [ ] H H l

Degenerate confguratons 3 4 4 H? H? 3 (case A) (case B) 4 3 Constrants: H,,3,4 Defne: en, H 4l * l, * ( ),, 3 ( l 4 ) 4 H 4 * H 4 4 k H * s rank- matr and tus not a omograp If H * s unque soluton, ten no omograp mappng (case B) If furter soluton H est, ten also αh * +βh (case A) (-D null-space n stead of -D null-space)

Solutons from lnes, etc. D omograpes from D lnes l H A l Mnmum of 4 lnes 3D Homograpes (5 dof) Mnmum of 5 ponts or 5 planes D affntes (6 dof) Mnmum of 3 ponts or lnes Conc provdes 5 constrants Med confguratons?

Cost functons Algebrac dstance Geometrc dstance Reprojecton error Comparson Geometrc nterpretaton Sampson error

DL mnmzes e A Algebrac dstance A resdual vector e partal vector for eac ( ) algebrac error vector d alg alg (,H ) e algebrac dstance (, ) a a ere a ( a, a, a3 ) (, H ) e A e d + d alg Not geometrcall/statstcall meanngfull, but gven good normalzaton t orks fne and s ver fast (use for ntalzaton)

ˆ Geometrc dstance measured coordnates estmated coordnates true coordnates d(.,.) Eucldean dstance (n mage) Error n one mage Ĥ argmn d H H (, H ) Smmetrc transfer error e.g. calbraton pattern ( - ), H + d(,h ) Ĥ argmn d Reprojecton error ( ) Ĥ, ˆ, ˆ argmn d(, ˆ ) + d(, ˆ ) H,ˆ,ˆ subject to ˆ Ĥˆ

Reprojecton error d ( - ), H + d(, H) d (, ˆ ) + d(, ˆ )

Comparson of geometrc and algebrac dstances e ˆ ˆ ˆ ˆ A Error n one mage ( ),, ( ) H ˆ, ˆ, ˆ ˆ 3 ( ) ( ) ( ) alg ˆ ˆ ˆ ˆ ˆ, d + ( ) ( ) ( ) ( ) ( ) d d + ˆ / ˆ, / ˆ / ˆ ˆ / ˆ / ˆ, alg / tpcal, but not, ecept for affntes ˆ 3 For affntes DL can mnmze geometrc dstance Possblt for teratve algortm

Geometrc nterpretaton of reprojecton error Estmatng omograp~ft surface to ponts (,,, ) n 4 H d ˆ ν H represents quadrcs n 4 (quadratc n ) ( ˆ ) + ( ˆ ) + ( ˆ ) + ( ˆ ) d (, ˆ ) + d(, ˆ ) (, ˆ ) + d(, ˆ ) d (, ν ) Analog to conc fttng H (,C) C d alg ( ),C d

Sampson error ˆ beteen algebrac and geometrc error ν H ˆ Vector tat mnmzes te geometrc error s te closest pont on te varet to te measurement Sampson error: st order appromaton of ˆ A CH ( ) C H C C H ( + δ ) CH( ) + δ + C H ( ) δ H δ Jδ ˆ C ( ˆ ) H e t J C H δ Fnd te vector tat mnmzes subject to δ Jδ e

δ Fnd te vector tat mnmzes subject to δ Jδ e Use Lagrange multplers: mnmze ( Jδ + e) δ δ - λ dervatves δ - λ J ( Jδ + e) δ J λ JJ λ + e λ JJ δ ( ) e ( ) J JJ e ˆ + e ( δ δ δ JJ ) e δ

Sampson error ˆ beteen algebrac and geometrc error ν H ˆ Vector tat mnmzes te geometrc error s te closest pont on te varet to te measurement Sampson error: st order appromaton of ˆ A CH ( ) C H C C H ( + δ ) CH ( ) + δ + C H ( ) δ H δ Jδ ˆ C ( ˆ ) H e δ Fnd te vector tat mnmzes subject to δ Jδ e δ ( ) JJ e δ δ e (Sampson error)

A fe ponts () () () (v) (v) (v) (v) Sampson appromaton δ e JJ For a D omograp (,,, ) e C H H J C ( ) s te algebrac error vector s a 4 matr, e.g. J ( ) e Smlar to algebrac error n fact, same as Maalanobs dstance Sampson error ndependent of lnear reparametrzaton (cancels out n beteen e and J) Must be summed for all ponts ( ) 3 3 + / + Close to geometrc error, but muc feer unknons e e e e e JJ ( ) JJ e

Statstcal cost functon and Mamum Lkelood Estmaton Optmal cost functon related to nose model Assume zero-mean sotropc Gaussan nose (assume outlers removed) Pr Pr Error n one mage (, ) / ( ) ( σ e d ) πσ d ({ } ) (,H ) ( ) / σ H e log Pr Π πσ ({ } H) d(,h ) + constant σ Mamum Lkelood Estmate (,H ) d

Statstcal cost functon and Mamum Lkelood Estmaton Optmal cost functon related to nose model Assume zero-mean sotropc Gaussan nose (assume outlers removed) Pr Pr ({ } H) (, ) / ( ) ( σ e d ) πσ Error n bot mages Π πσ e (, ) d (,H ) /( σ ) d + Mamum Lkelood Estmate (, ˆ ) + d(, ˆ ) d

Maalanobs dstance General Gaussan case Measurement t covarance matr Σ ( ) ( ) Σ Σ Σ Σ + Error n to mages (ndependent) Σ Σ + Varng covarances

Net class: Parameter estmaton (contnued) ransformaton nvarance and normalzaton Iteratve mnmzaton Robust estmaton

Upcomng assgnment ake to or more potograps taken from a sngle vepont Compute panorama Use dfferent measures DL, MLE Use Matlab Due Feb. 3