Theory of Bouguet s MatLab Camera Calibration Toolbox

Size: px
Start display at page:

Download "Theory of Bouguet s MatLab Camera Calibration Toolbox"

Transcription

1 Theory of Bouguet s MatLab Camera Calibration Toolbox Yuji Oyamada 1 HVRL, University 2 Chair for Computer Aided Medical Procedure (CAMP) Technische Universität München June 26, 2012

2 MatLab Camera Calibration Toolbox Calibration using a planar calibration object. Points correspondence requires manual click on object s corners. Core calibration part is fully automatic. C implementation is included in OpenCV. Web

3 Functions Extract grid corners: Points correspondence Calibration: Core calibration Show Extrinsic: Visualization of estimated extrinsic parameters Reproject on images: Visualization of reprojection error Analyse error: Recomp. corners: Recompute the corners using estimated parameters

4 Input: Extract grid corners n sq {x,y}: Number of square along {x,y} axis d{x,y}: Size of square in m unit. I: Input image Output: {x i}, i = 1,..., N: Set of detected 2D points. {X i}, i = 1,..., N: Set of known 3D points. where x i and X i have M i elements as [ ] x i(1) x i(j) x i(mi ) x i = and y i(1) y i(j) y i(m i ) X i(1) X i(j) X i(m i ) X i = Y i(1) Y i(j) Y i(m i ) Z i(1) Z i(j) Z i(m i )

5 Calibration Input: {x i}: Set of detected 2D points. {X i}: Set of known 3D points. Output: Intrinsic parameters Extrinsic parameters Method: 1. Initialize intrinsic and extrinsic params. 2. Non-linear optimization to refine all parameters.

6 Camera parameters Intrinsic parameters: fc R 2 : Camera focal length cc R 2 : Principal point coordinates alpha c R 1 : Skew coefficient kc R 5 : Distortion coefficients KK R 3 3 : The camera matrix (containing fc, cc, alpha c) Extrinsic parameters for i-th image: omc i R 3 : Rotation angles Tc i R 3 : Translation vectors Rc i R 3 3 : Rotation matrix computed as Rc i = rodrigues(omc i)

7 Initialization 1/2: Intrinsic parameters Input {x i}: Set of detected 2D points. {X i}: Set of known 3D points. Output Initialized intrinsic parameters Initialized extrinsic parameters

8 Initialization 1/2: Intrinsic parameters 1. Roughly compute intrinsic params. 2. Refine intrinsic params. 3. Compute extrinsic params. Step1: fc: Computed based on all the vanishing points. cc: Center of image. alpha c = 0. kc = 0.

9 Initialization 1/2: Intrinsic parameters Step2: Given N sets of corresponding points {x k, X k }, 1. Compute Homography {H k }. 2. Build linear equations Vb = 0 and solve the equations Detail. 3. Compute A from estimated b. v 1,12 (v 1,11 v 1,22 ) v k,12 V (v k,11 v k,22 ) v N,12 (v N,11 v N,22 ) v k,ij = [h k,i1 h k,j1, h k,i1 h k,j2 + h k,i2 h k,j1, h k,i2 h k,j2, h k,i3 h k,j1 + h k,i1 h k,j3, h k,i3 h k,j2 + h k,i2 h k,j3, h k,i3 h k,j3 ]

10 Initialization 2/2: Extrinsic parameters Step3: Compute each extrinsic parameters from estimated A and pre-computed H k as where λ k = r k,1 = λ k A 1 h k,1 r k,2 = λ k A 1 h k,2 r k,3 = r k,1 r k,2 t k = λ k A 1 h k,3 1 A 1 h k,1 = 1 A 1 h k,2

11 Solves the following equations: Non-linear optimization ˆp = arg min p x f (p) 2 2 where p = {fc, cc, alpha c, kc, {omc i}, {Tc i}} and the function f projects the known 3D points X onto 2D image plane with the argument p. Sparse Levenberg-Marquardt algorithm iteratively updates ˆp with an initial estimate ˆp 0 as while( x f (ˆp i 1 ) 2 2 > ɛ) ˆp i = ˆp i 1 + p i ; i = i + 1;

12 Given Image projection x p = (x p, y p, 1) : Pixel coordinate of a point on the image plane. x d = (x d, y d, 1) : Normalized coordinate distorted by lens. x n = (x n, y n, 1) : Normalized pinhole image projection. k R 5 : Lens distortion parameters. KK R 3 3 : Camera calibration matrix. where fc(1) alpha c(1) cc(1) KK = 0 fc(2) cc(2) x p = KKx d = KK(cdistx n + delta x)

13 Given Lens distortion x n = x: Normalized pinhole image projection. k R 5 : Lens distortion parameters. Compute x d = xd: Normalized coordinate distorted by lens. cdist: Radial distortion delta x: Tangential distortion xd = cdist x + delta x [ ] cdist x + delta x(1) = cdist y + delta x(2)

14 Radial distortion where xd1 = [ ] x cdist y cdist cdist = 1 + k(1) r2 + k(2) r4 + k(5) r6 r2 = x 2 i + y 2 i, r4 = r2 2, r6 = r2 3

15 Tangential distortion where delta x = [ ] k(3) a1 + k(4) a2 k(3) a3 + k(4) a1 a1 = 2x i y i a2 = r2 + 2x 2 i a3 = r2 + 2y 2 i

16 Closed form solution for initialization X u s v = A [ r 1 r 2 r 3 t ] Y 0 = A [ r 1 r 2 t ] X Y X = H Y 1 Since r 1 and r 2 are orthonormal, we have following two constraints: h 1 A A 1 h 2 = 0 h 1 A A 1 h 1 = h 2 A A 1 h 2, where A A 1 describes the image of the absolute conic.

17 Closed form solution for initialization Let B 11 B 21 B 31 B = A A 1 B 12 B 22 B 32 B 13 B 23 B 33 = 1 α 2 γ α 2 β γ α 2 β v 0 γ u 0 β α 2 β γ γ(v 0γ u 0 β) v α 2 β 2 β 2 α 2 β 2 0 β 2 v 0 γ u 0 β α 2 β γ(v 0γ u 0 β) α 2 β 2 v 0 β 2 (v 0 γ u 0 β) 2 α 2 β 2 + v 2 0 β Note that B is symmetric, defined by a 6D vector b [ B 11, B 12, B 22, B 13, B 23, B 3 ]

18 Closed form solution for initialization Let the i-th column vector of H be h i = [ h i1, h i2, h i3 ]. Then, we have following linear equation where h i Bh j = v ij b v ij = [h i1 h j1, h i1 h j2 + h i2 h j1, h i2 h j2 h i3 h j1 + h i1 h j3, h i3 h j2 + h i2 h j3, h i3 h j3 ]

19 Closed form solution for initialization The above two constraints can be rewritten as two homogeneous equation w.r.t. unknown b as [ v ] 12 (v 11 v 22 ) b = 0 Given n images, we have Vb = 0, where V R 2n 6.

20 Least Squares Let f ( ) a function projecting 3D points X onto 2D points x with projection parameters p as bmx = f (p) = px. Our task is to find an optimal p such that minimizes x px 2 2. In the case of non-linear function f ( ), we iteratively update the estimate ˆp with an initial estimate ˆp 0 as while( x px 2 2 > ɛ) ˆp i = ˆp i 1 + p i ; i = i + 1;

21 Least Squares: Newton s method Given a previous estimate ˆp i, estimate an update i s.t. i = arg min f (ˆp i + ) X 2 2 = arg min f (ˆp i) + J X 2 2 = arg min ɛ i + J 2 2 where J = f p. Then, solve the normal equations J J = J ɛ i

22 Levenberg-Marquardt (LM) iteration The normal equation is replaced by the augmented normal equations as J J = J ɛ i (J J + λi) = J ɛ i where I denotes the identity matrix. An initial value of λ is 10 3 times the average of the diagonal elements of N = J J.

23 Sparse LM LM is suitable for minimization w.r.t. a small number of parameters. The central step of LM, solving the normal equations, has complexity N 3 in the number of parameters and is repeated many times. The normal equation matrix has a certain sparse block structure.

24 Sparse LM Let p R M be the parameter vector that is able to be partitioned into parameter vectors as p = (a, b ). Given a measurement vector x R N Let x be the covariance matrix for the measurement vector. A general function f : R M R N takes p to the estimated measurement vector ˆx = f (p). ɛ denotes the difference x ˆx between the measured and the estimated vectors.

25 Sparse LM The set of equations Jδ = ɛ solved as the central step in the LM has the form ( ) δa Jδ = [A B] = ɛ. Then, the normal equations J 1 x at each step of LM are of the form [ A 1 x A A ] 1 ( ) ( x B B 1 x A B δa 1 = x B δ b δ b Jδ = J 1 x ɛ to be solved A 1 x B 1 x ) ɛ ɛ

26 Let Sparse LM U = A 1 x A W = A 1 x B V = B 1 x B and denotes augmented matrix by λ. The normal equations are rewritten as [ U ] ( ) ( ) W δa ɛa = W V δ b ɛ B [ U WV 1 W ] ( ) ( 0 δa ɛa WV 1 ) ɛ = B W This results in the elimination of the top right hand block. V δ b ɛ B

27 Sparse LM The top half of this set of equations is (U WV 1 W )δ a = ɛ A WV 1 ɛ B Subsequently, the value of δ a may be found by back-substitution, giving V δ b = ɛ B W δ a

28 Sparse LM p = (a, b ), where a = (fc, cc, alpha c, kc ) and b = ({omc i Tc i }) The Jacobian matrix is where [ ˆx ˆx a = ˆx b = [ fc, ˆx cc, ˆx omc 1, J = ˆx p = ˆx alpha c, ˆx ] kc ˆx Tc 1,, [ ˆx a, ˆx ] = [A, B] b ˆx omc i ˆx Tc i, ˆx omc N ] ˆx Tc N

29 Sparse LM The normal equation is rewritten as ([ ] ) [ ] A [A ] A B + λi p = ɛ x B N A i A i i=1 N B i A i i=1 N i=1 N i=1 B A i B i [ B + λi A ] ɛ p = x B ɛ x i B i

30 Sparse LM J J = N A i A i A 1 B 1 A i B i A N B N i=1 B 1 A 1 B 1 B 1. B i A i. B N A N... B i B i... B N B N

31 Sparse LM J ɛ x = N A i ɛ x i=1 B 1 ɛ x. B i ɛ x. B N ɛ x

32 Sparse LM When each image has different number of corresponding points (M i M j, if i j), each A i and B i have different size as

33 Sparse LM However, the difference does not matter because A A R d int d int A B R d int d ex B A R dex d int B B R dex dex where d int denotes dimension of intrinsic params and d ex denotes dimension of extrinsic params.

Sparse Levenberg-Marquardt algorithm.

Sparse Levenberg-Marquardt algorithm. Sparse Levenberg-Marquardt algorithm. R. I. Hartley and A. Zisserman: Multiple View Geometry in Computer Vision. Cambridge University Press, second edition, 2004. Appendix 6 was used in part. The Levenberg-Marquardt

More information

Pose Tracking II! Gordon Wetzstein! Stanford University! EE 267 Virtual Reality! Lecture 12! stanford.edu/class/ee267/!

Pose Tracking II! Gordon Wetzstein! Stanford University! EE 267 Virtual Reality! Lecture 12! stanford.edu/class/ee267/! Pose Tracking II! Gordon Wetzstein! Stanford University! EE 267 Virtual Reality! Lecture 12! stanford.edu/class/ee267/!! WARNING! this class will be dense! will learn how to use nonlinear optimization

More information

3D from Photographs: Camera Calibration. Dr Francesco Banterle

3D from Photographs: Camera Calibration. Dr Francesco Banterle 3D from Photographs: Camera Calibration Dr Francesco Banterle francesco.banterle@isti.cnr.it 3D from Photographs Automatic Matching of Images Camera Calibration Photographs Surface Reconstruction Dense

More information

Camera calibration by Zhang

Camera calibration by Zhang Camera calibration by Zhang Siniša Kolarić September 2006 Abstract In this presentation, I present a way to calibrate a camera using the method by Zhang. NOTE. This

More information

Augmented Reality VU Camera Registration. Prof. Vincent Lepetit

Augmented Reality VU Camera Registration. Prof. Vincent Lepetit Augmented Reality VU Camera Registration Prof. Vincent Lepetit Different Approaches to Vision-based 3D Tracking [From D. Wagner] [From Drummond PAMI02] [From Davison ICCV01] Consider natural features Consider

More information

Camera calibration. Outline. Pinhole camera. Camera projection models. Nonlinear least square methods A camera calibration tool

Camera calibration. Outline. Pinhole camera. Camera projection models. Nonlinear least square methods A camera calibration tool Outline Camera calibration Camera projection models Camera calibration i Nonlinear least square methods A camera calibration tool Applications Digital Visual Effects Yung-Yu Chuang with slides b Richard

More information

Visual SLAM Tutorial: Bundle Adjustment

Visual SLAM Tutorial: Bundle Adjustment Visual SLAM Tutorial: Bundle Adjustment Frank Dellaert June 27, 2014 1 Minimizing Re-projection Error in Two Views In a two-view setting, we are interested in finding the most likely camera poses T1 w

More information

Camera Calibration The purpose of camera calibration is to determine the intrinsic camera parameters (c 0,r 0 ), f, s x, s y, skew parameter (s =

Camera Calibration The purpose of camera calibration is to determine the intrinsic camera parameters (c 0,r 0 ), f, s x, s y, skew parameter (s = Camera Calibration The purpose of camera calibration is to determine the intrinsic camera parameters (c 0,r 0 ), f, s x, s y, skew parameter (s = cotα), and the lens distortion (radial distortion coefficient

More information

Pose estimation from point and line correspondences

Pose estimation from point and line correspondences Pose estimation from point and line correspondences Giorgio Panin October 17, 008 1 Problem formulation Estimate (in a LSE sense) the pose of an object from N correspondences between known object points

More information

Homographies and Estimating Extrinsics

Homographies and Estimating Extrinsics Homographies and Estimating Extrinsics Instructor - Simon Lucey 16-423 - Designing Computer Vision Apps Adapted from: Computer vision: models, learning and inference. Simon J.D. Prince Review: Motivation

More information

nonrobust estimation The n measurement vectors taken together give the vector X R N. The unknown parameter vector is P R M.

nonrobust estimation The n measurement vectors taken together give the vector X R N. The unknown parameter vector is P R M. Introduction to nonlinear LS estimation R. I. Hartley and A. Zisserman: Multiple View Geometry in Computer Vision. Cambridge University Press, 2ed., 2004. After Chapter 5 and Appendix 6. We will use x

More information

OPPA European Social Fund Prague & EU: We invest in your future.

OPPA European Social Fund Prague & EU: We invest in your future. OPPA European Social Fund Prague & EU: We invest in your future. 1D Projective Coordinates The 1-D projective coordinate of a point P : [P ] = [P P 0 P I P ] = [p p 0 p I p] = p pi p 0 p I p 0 p p p p

More information

Linear and Nonlinear Optimization

Linear and Nonlinear Optimization Linear and Nonlinear Optimization German University in Cairo October 10, 2016 Outline Introduction Gradient descent method Gauss-Newton method Levenberg-Marquardt method Case study: Straight lines have

More information

Camera calibration Triangulation

Camera calibration Triangulation Triangulation Perspective projection in homogenous coordinates ~x img I 0 apple R t 0 T 1 ~x w ~x img R t ~x w Matrix transformations in 2D ~x img K R t ~x w K = 2 3 1 0 t u 40 1 t v 5 0 0 1 Translation

More information

EE Camera & Image Formation

EE Camera & Image Formation Electric Electronic Engineering Bogazici University February 21, 2018 Introduction Introduction Camera models Goal: To understand the image acquisition process. Function of the camera Similar to that of

More information

13. Nonlinear least squares

13. Nonlinear least squares L. Vandenberghe ECE133A (Fall 2018) 13. Nonlinear least squares definition and examples derivatives and optimality condition Gauss Newton method Levenberg Marquardt method 13.1 Nonlinear least squares

More information

Chapter 4 Systems of Linear Equations; Matrices

Chapter 4 Systems of Linear Equations; Matrices Chapter 4 Systems of Linear Equations; Matrices Section 5 Inverse of a Square Matrix Learning Objectives for Section 4.5 Inverse of a Square Matrix The student will be able to identify identity matrices

More information

Numerical computation II. Reprojection error Bundle adjustment Family of Newtonʼs methods Statistical background Maximum likelihood estimation

Numerical computation II. Reprojection error Bundle adjustment Family of Newtonʼs methods Statistical background Maximum likelihood estimation Numerical computation II Reprojection error Bundle adjustment Family of Newtonʼs methods Statistical background Maximum likelihood estimation Reprojection error Reprojection error = Distance between the

More information

Lecture 3: QR-Factorization

Lecture 3: QR-Factorization Lecture 3: QR-Factorization This lecture introduces the Gram Schmidt orthonormalization process and the associated QR-factorization of matrices It also outlines some applications of this factorization

More information

Augmented Reality VU numerical optimization algorithms. Prof. Vincent Lepetit

Augmented Reality VU numerical optimization algorithms. Prof. Vincent Lepetit Augmented Reality VU numerical optimization algorithms Prof. Vincent Lepetit P3P: can exploit only 3 correspondences; DLT: difficult to exploit the knowledge of the internal parameters correctly, and the

More information

Applied Mathematics 205. Unit I: Data Fitting. Lecturer: Dr. David Knezevic

Applied Mathematics 205. Unit I: Data Fitting. Lecturer: Dr. David Knezevic Applied Mathematics 205 Unit I: Data Fitting Lecturer: Dr. David Knezevic Unit I: Data Fitting Chapter I.4: Nonlinear Least Squares 2 / 25 Nonlinear Least Squares So far we have looked at finding a best

More information

16. Analysis and Computation of the Intrinsic Camera Parameters

16. Analysis and Computation of the Intrinsic Camera Parameters 16. Analysis and Computation of the Intrinsic Camera Parameters Eduardo Bayro-Corrochano and Bodo Rosenhahn Institute of Computer Science and Applied Mathematics, Christian-Albrechts-University of Kiel

More information

Quality Report Generated with Pix4Dmapper Pro version

Quality Report Generated with Pix4Dmapper Pro version Quality Report Generated with Pix4Dmapper Pro version 3.0.13 Important: Click on the different icons for: Help to analyze the results in the Quality Report Additional information about the sections Click

More information

Method 1: Geometric Error Optimization

Method 1: Geometric Error Optimization Method 1: Geometric Error Optimization we need to encode the constraints ŷ i F ˆx i = 0, rank F = 2 idea: reconstruct 3D point via equivalent projection matrices and use reprojection error equivalent projection

More information

ANALYTICAL IMAGE FEATURES

ANALYTICAL IMAGE FEATURES Chapter 6 ANALYTICAL IMAGE FEATURES Chapter laid the geometric foundations of image formation. This chapter uses analytical geometry to quantify more precisely the relationship between a camera, the objects

More information

Lecture 4.3 Estimating homographies from feature correspondences. Thomas Opsahl

Lecture 4.3 Estimating homographies from feature correspondences. Thomas Opsahl Lecture 4.3 Estimating homographies from feature correspondences Thomas Opsahl Homographies induced by central projection 1 H 2 1 H 2 u uu 2 3 1 Homography Hu = u H = h 1 h 2 h 3 h 4 h 5 h 6 h 7 h 8 h

More information

Camera Calibration. (Trucco, Chapter 6) -Toproduce an estimate of the extrinsic and intrinsic camera parameters.

Camera Calibration. (Trucco, Chapter 6) -Toproduce an estimate of the extrinsic and intrinsic camera parameters. Camera Calibration (Trucco, Chapter 6) What is the goal of camera calibration? -Toproduce an estimate of the extrinsic and intrinsic camera parameters. Procedure -Given the correspondences beteen a set

More information

A Stellar Gyroscope for CubeSat Attitude Determination

A Stellar Gyroscope for CubeSat Attitude Determination A Stellar Gyroscope for CubeSat Attitude Determination Samir A. Rawashdeh and James E. Lumpp, Jr. Space Systems Laboratory University of Kentucky James Barrington-Brown and Massimiliano Pastena SSBV Space

More information

Nonlinear Least Squares

Nonlinear Least Squares Nonlinear Least Squares Stephen Boyd EE103 Stanford University December 6, 2016 Outline Nonlinear equations and least squares Examples Levenberg-Marquardt algorithm Nonlinear least squares classification

More information

Structure from Motion. CS4670/CS Kevin Matzen - April 15, 2016

Structure from Motion. CS4670/CS Kevin Matzen - April 15, 2016 Structure from Motion CS4670/CS5670 - Kevin Matzen - April 15, 2016 Video credit: Agarwal, et. al. Building Rome in a Day, ICCV 2009 Roadmap What we ve seen so far Single view modeling (1 camera) Stereo

More information

A Study of Kruppa s Equation for Camera Self-calibration

A Study of Kruppa s Equation for Camera Self-calibration Proceedings of the International Conference of Machine Vision and Machine Learning Prague, Czech Republic, August 14-15, 2014 Paper No. 57 A Study of Kruppa s Equation for Camera Self-calibration Luh Prapitasari,

More information

Practical rotation angle measurement method by monocular vision

Practical rotation angle measurement method by monocular vision Practical rotation angle measurement method by monocular vision Hongxin Dong,, * Qiang Fu, Xu Zhao, Quan Quan,,2 and Ruifeng Zhang School of Automation Science and Electrical Engineering, Beihang University,

More information

Deuxième examen partiel A 2013

Deuxième examen partiel A 2013 12 décembre 2013 13h30 à 16h20 PLT-2501 Deuxième examen partiel A 2013 Toute documentation permise sauf Internet QUESTION 1 (15 points au total) Camera calibration. When a camera is calibrated, a target

More information

CS181 Midterm 2 Practice Solutions

CS181 Midterm 2 Practice Solutions CS181 Midterm 2 Practice Solutions 1. Convergence of -Means Consider Lloyd s algorithm for finding a -Means clustering of N data, i.e., minimizing the distortion measure objective function J({r n } N n=1,

More information

Principal Component Analysis (PCA) for Sparse High-Dimensional Data

Principal Component Analysis (PCA) for Sparse High-Dimensional Data AB Principal Component Analysis (PCA) for Sparse High-Dimensional Data Tapani Raiko, Alexander Ilin, and Juha Karhunen Helsinki University of Technology, Finland Adaptive Informatics Research Center Principal

More information

A Theory of Multi-Layer Flat Refractive Geometry Supplementary Materials

A Theory of Multi-Layer Flat Refractive Geometry Supplementary Materials A Theory of Multi-Layer Flat Refractive Geometry Supplementary Materials Amit Agrawal, Srikumar Ramalingam, Yuichi Taguchi Mitsubishi Electric Research Labs MERL) [agrawal,ramalingam,taguchi] at merl.com

More information

Introduction to pinhole cameras

Introduction to pinhole cameras Introduction to pinhole cameras Lesson given by Sébastien Piérard in the course Introduction aux techniques audio et vidéo (ULg, Pr JJ Embrechts) INTELSIG, Montefiore Institute, University of Liège, Belgium

More information

CS4495/6495 Introduction to Computer Vision. 3D-L3 Fundamental matrix

CS4495/6495 Introduction to Computer Vision. 3D-L3 Fundamental matrix CS4495/6495 Introduction to Computer Vision 3D-L3 Fundamental matrix Weak calibration Main idea: Estimate epipolar geometry from a (redundant) set of point correspondences between two uncalibrated cameras

More information

Self-Calibration of a Moving Camera by Pre-Calibration

Self-Calibration of a Moving Camera by Pre-Calibration Self-Calibration of a Moving Camera by Pre-Calibration Peter Sturm To cite this version: Peter Sturm. Self-Calibration of a Moving Camera by Pre-Calibration. Robert B. Fisher and Emanuele Trucco. 7th British

More information

EPIPOLAR GEOMETRY WITH MANY DETAILS

EPIPOLAR GEOMETRY WITH MANY DETAILS EPIPOLAR GEOMERY WIH MANY DEAILS hank ou for the slides. he come mostl from the following source. Marc Pollefes U. of North Carolina hree questions: (i) Correspondence geometr: Given an image point in

More information

CS 231A Computer Vision (Fall 2011) Problem Set 2

CS 231A Computer Vision (Fall 2011) Problem Set 2 CS 231A Computer Vision (Fall 2011) Problem Set 2 Solution Set Due: Oct. 28 th, 2011 (5pm) 1 Some Projective Geometry Problems (15 points) Suppose there are two parallel lines that extend to infinity in

More information

Optimisation on Manifolds

Optimisation on Manifolds Optimisation on Manifolds K. Hüper MPI Tübingen & Univ. Würzburg K. Hüper (MPI Tübingen & Univ. Würzburg) Applications in Computer Vision Grenoble 18/9/08 1 / 29 Contents 2 Examples Essential matrix estimation

More information

Least squares problems

Least squares problems Least squares problems How to state and solve them, then evaluate their solutions Stéphane Mottelet Université de Technologie de Compiègne 30 septembre 2016 Stéphane Mottelet (UTC) Least squares 1 / 55

More information

Uncertainty Models in Quasiconvex Optimization for Geometric Reconstruction

Uncertainty Models in Quasiconvex Optimization for Geometric Reconstruction Uncertainty Models in Quasiconvex Optimization for Geometric Reconstruction Qifa Ke and Takeo Kanade Department of Computer Science, Carnegie Mellon University Email: ke@cmu.edu, tk@cs.cmu.edu Abstract

More information

3D Computer Vision - WT 2004

3D Computer Vision - WT 2004 3D Computer Vision - WT 2004 Singular Value Decomposition Darko Zikic CAMP - Chair for Computer Aided Medical Procedures November 4, 2004 1 2 3 4 5 Properties For any given matrix A R m n there exists

More information

L3: Review of linear algebra and MATLAB

L3: Review of linear algebra and MATLAB L3: Review of linear algebra and MATLAB Vector and matrix notation Vectors Matrices Vector spaces Linear transformations Eigenvalues and eigenvectors MATLAB primer CSCE 666 Pattern Analysis Ricardo Gutierrez-Osuna

More information

Vision par ordinateur

Vision par ordinateur Vision par ordinateur Géométrie épipolaire Frédéric Devernay Avec des transparents de Marc Pollefeys Epipolar geometry π Underlying structure in set of matches for rigid scenes C1 m1 l1 M L2 L1 l T 1 l

More information

Linear Algebra: Matrix Eigenvalue Problems

Linear Algebra: Matrix Eigenvalue Problems CHAPTER8 Linear Algebra: Matrix Eigenvalue Problems Chapter 8 p1 A matrix eigenvalue problem considers the vector equation (1) Ax = λx. 8.0 Linear Algebra: Matrix Eigenvalue Problems Here A is a given

More information

The Multibody Trifocal Tensor: Motion Segmentation from 3 Perspective Views

The Multibody Trifocal Tensor: Motion Segmentation from 3 Perspective Views The Multibody Trifocal Tensor: Motion Segmentation from 3 Perspective Views Richard Hartley 1,2 and RenéVidal 2,3 1 Dept. of Systems Engineering 3 Center for Imaging Science Australian National University

More information

Math 273a: Optimization Netwon s methods

Math 273a: Optimization Netwon s methods Math 273a: Optimization Netwon s methods Instructor: Wotao Yin Department of Mathematics, UCLA Fall 2015 some material taken from Chong-Zak, 4th Ed. Main features of Newton s method Uses both first derivatives

More information

TRACKING and DETECTION in COMPUTER VISION

TRACKING and DETECTION in COMPUTER VISION Technischen Universität München Winter Semester 2013/2014 TRACKING and DETECTION in COMPUTER VISION Template tracking methods Slobodan Ilić Template based-tracking Energy-based methods The Lucas-Kanade(LK)

More information

Least-Squares Performance of Analog Product Codes

Least-Squares Performance of Analog Product Codes Copyright 004 IEEE Published in the Proceedings of the Asilomar Conference on Signals, Systems and Computers, 7-0 ovember 004, Pacific Grove, California, USA Least-Squares Performance of Analog Product

More information

Least-Squares Fitting of Model Parameters to Experimental Data

Least-Squares Fitting of Model Parameters to Experimental Data Least-Squares Fitting of Model Parameters to Experimental Data Div. of Mathematical Sciences, Dept of Engineering Sciences and Mathematics, LTU, room E193 Outline of talk What about Science and Scientific

More information

Finite Element Methods with Linear B-Splines

Finite Element Methods with Linear B-Splines Finite Element Methods with Linear B-Splines K. Höllig, J. Hörner and M. Pfeil Universität Stuttgart Institut für Mathematische Methoden in den Ingenieurwissenschaften, Numerik und geometrische Modellierung

More information

Visual Object Recognition

Visual Object Recognition Visual Object Recognition Lecture 2: Image Formation Per-Erik Forssén, docent Computer Vision Laboratory Department of Electrical Engineering Linköping University Lecture 2: Image Formation Pin-hole, and

More information

CSE 160 Lecture 13. Numerical Linear Algebra

CSE 160 Lecture 13. Numerical Linear Algebra CSE 16 Lecture 13 Numerical Linear Algebra Announcements Section will be held on Friday as announced on Moodle Midterm Return 213 Scott B Baden / CSE 16 / Fall 213 2 Today s lecture Gaussian Elimination

More information

Numerical Programming I (for CSE)

Numerical Programming I (for CSE) Technische Universität München WT 1/13 Fakultät für Mathematik Prof. Dr. M. Mehl B. Gatzhammer January 1, 13 Numerical Programming I (for CSE) Tutorial 1: Iterative Methods 1) Relaxation Methods a) Let

More information

Non-polynomial Least-squares fitting

Non-polynomial Least-squares fitting Applied Math 205 Last time: piecewise polynomial interpolation, least-squares fitting Today: underdetermined least squares, nonlinear least squares Homework 1 (and subsequent homeworks) have several parts

More information

Homogeneous Coordinates

Homogeneous Coordinates Homogeneous Coordinates Basilio Bona DAUIN-Politecnico di Torino October 2013 Basilio Bona (DAUIN-Politecnico di Torino) Homogeneous Coordinates October 2013 1 / 32 Introduction Homogeneous coordinates

More information

Bundle Adjustment for 3-D Reconstruction: Implementation and Evaluation

Bundle Adjustment for 3-D Reconstruction: Implementation and Evaluation Memoirs of the Faculty of Engineering Okayama University Vol 45 pp 1 9 January 2011 Bundle Adjustment for 3-D Reconstruction: Implementation and Evaluation Kenichi KANATANI Department of Computer Science

More information

1 EM algorithm: updating the mixing proportions {π k } ik are the posterior probabilities at the qth iteration of EM.

1 EM algorithm: updating the mixing proportions {π k } ik are the posterior probabilities at the qth iteration of EM. Université du Sud Toulon - Var Master Informatique Probabilistic Learning and Data Analysis TD: Model-based clustering by Faicel CHAMROUKHI Solution The aim of this practical wor is to show how the Classification

More information

Linear algebra for computational statistics

Linear algebra for computational statistics University of Seoul May 3, 2018 Vector and Matrix Notation Denote 2-dimensional data array (n p matrix) by X. Denote the element in the ith row and the jth column of X by x ij or (X) ij. Denote by X j

More information

Geometric Image Manipulation

Geometric Image Manipulation Bruce A. Draper J. Ross Beveridge, January 24, 204 Geometric Image Manipulation Lecture (part a) January 24, 204 Bruce A. Draper J. Ross Beveridge, January 24, 204 Status Update Programming assignment

More information

1 Curvature of submanifolds of Euclidean space

1 Curvature of submanifolds of Euclidean space Curvature of submanifolds of Euclidean space by Min Ru, University of Houston 1 Curvature of submanifolds of Euclidean space Submanifold in R N : A C k submanifold M of dimension n in R N means that for

More information

Position & motion in 2D and 3D

Position & motion in 2D and 3D Position & motion in 2D and 3D! METR4202: Guest lecture 22 October 2014 Peter Corke Peter Corke Robotics, Vision and Control FUNDAMENTAL ALGORITHMS IN MATLAB 123 Sections 11.1, 15.2 Queensland University

More information

CSE 252B: Computer Vision II

CSE 252B: Computer Vision II CSE 252B: Computer Vision II Lecturer: Serge Belongie Scribe: Hamed Masnadi Shirazi, Solmaz Alipour LECTURE 5 Relationships between the Homography and the Essential Matrix 5.1. Introduction In practice,

More information

Structure from Motion. Read Chapter 7 in Szeliski s book

Structure from Motion. Read Chapter 7 in Szeliski s book Structure from Motion Read Chapter 7 in Szeliski s book Schedule (tentative) 2 # date topic 1 Sep.22 Introduction and geometry 2 Sep.29 Invariant features 3 Oct.6 Camera models and calibration 4 Oct.13

More information

Probabilistic & Unsupervised Learning

Probabilistic & Unsupervised Learning Probabilistic & Unsupervised Learning Week 2: Latent Variable Models Maneesh Sahani maneesh@gatsby.ucl.ac.uk Gatsby Computational Neuroscience Unit, and MSc ML/CSML, Dept Computer Science University College

More information

Linear Algebra & Geometry why is linear algebra useful in computer vision?

Linear Algebra & Geometry why is linear algebra useful in computer vision? Linear Algebra & Geometry why is linear algebra useful in computer vision? References: -Any book on linear algebra! -[HZ] chapters 2, 4 Some of the slides in this lecture are courtesy to Prof. Octavia

More information

Vision 3D articielle Session 2: Essential and fundamental matrices, their computation, RANSAC algorithm

Vision 3D articielle Session 2: Essential and fundamental matrices, their computation, RANSAC algorithm Vision 3D articielle Session 2: Essential and fundamental matrices, their computation, RANSAC algorithm Pascal Monasse monasse@imagine.enpc.fr IMAGINE, École des Ponts ParisTech Contents Some useful rules

More information

Kadath: a spectral solver and its application to black hole spacetimes

Kadath: a spectral solver and its application to black hole spacetimes Kadath: a spectral solver and its application to black hole spacetimes Philippe Grandclément Laboratoire de l Univers et de ses Théories (LUTH) CNRS / Observatoire de Paris F-92195 Meudon, France philippe.grandclement@obspm.fr

More information

Inverse Theory. COST WaVaCS Winterschool Venice, February Stefan Buehler Luleå University of Technology Kiruna

Inverse Theory. COST WaVaCS Winterschool Venice, February Stefan Buehler Luleå University of Technology Kiruna Inverse Theory COST WaVaCS Winterschool Venice, February 2011 Stefan Buehler Luleå University of Technology Kiruna Overview Inversion 1 The Inverse Problem 2 Simple Minded Approach (Matrix Inversion) 3

More information

Matrix Derivatives and Descent Optimization Methods

Matrix Derivatives and Descent Optimization Methods Matrix Derivatives and Descent Optimization Methods 1 Qiang Ning Department of Electrical and Computer Engineering Beckman Institute for Advanced Science and Techonology University of Illinois at Urbana-Champaign

More information

Class Averaging in Cryo-Electron Microscopy

Class Averaging in Cryo-Electron Microscopy Class Averaging in Cryo-Electron Microscopy Zhizhen Jane Zhao Courant Institute of Mathematical Sciences, NYU Mathematics in Data Sciences ICERM, Brown University July 30 2015 Single Particle Reconstruction

More information

Feature Tracking. 2/27/12 ECEn 631

Feature Tracking. 2/27/12 ECEn 631 Corner Extraction Feature Tracking Mostly for multi-frame applications Object Tracking Motion detection Image matching Image mosaicing 3D modeling Object recognition Homography estimation... Global Features

More information

Trust-region methods for rectangular systems of nonlinear equations

Trust-region methods for rectangular systems of nonlinear equations Trust-region methods for rectangular systems of nonlinear equations Margherita Porcelli Dipartimento di Matematica U.Dini Università degli Studi di Firenze Joint work with Maria Macconi and Benedetta Morini

More information

A geometric interpretation of the homogeneous coordinates is given in the following Figure.

A geometric interpretation of the homogeneous coordinates is given in the following Figure. Introduction Homogeneous coordinates are an augmented representation of points and lines in R n spaces, embedding them in R n+1, hence using n + 1 parameters. This representation is useful in dealing with

More information

Basic Math for

Basic Math for Basic Math for 16-720 August 23, 2002 1 Linear Algebra 1.1 Vectors and Matrices First, a reminder of a few basic notations, definitions, and terminology: Unless indicated otherwise, vectors are always

More information

Single view metrology

Single view metrology EECS 44 Computer vision Single view metrology Review calibration Lines and planes at infinity Absolute conic Estimating geometry from a single image Etensions Reading: [HZ] Chapters,3,8 Calibration Problem

More information

Computational Methods. Least Squares Approximation/Optimization

Computational Methods. Least Squares Approximation/Optimization Computational Methods Least Squares Approximation/Optimization Manfred Huber 2011 1 Least Squares Least squares methods are aimed at finding approximate solutions when no precise solution exists Find the

More information

Lecture 2: Linear Algebra Review

Lecture 2: Linear Algebra Review EE 227A: Convex Optimization and Applications January 19 Lecture 2: Linear Algebra Review Lecturer: Mert Pilanci Reading assignment: Appendix C of BV. Sections 2-6 of the web textbook 1 2.1 Vectors 2.1.1

More information

Parallel Scientific Computing

Parallel Scientific Computing IV-1 Parallel Scientific Computing Matrix-vector multiplication. Matrix-matrix multiplication. Direct method for solving a linear equation. Gaussian Elimination. Iterative method for solving a linear equation.

More information

April 26, Applied mathematics PhD candidate, physics MA UC Berkeley. Lecture 4/26/2013. Jed Duersch. Spd matrices. Cholesky decomposition

April 26, Applied mathematics PhD candidate, physics MA UC Berkeley. Lecture 4/26/2013. Jed Duersch. Spd matrices. Cholesky decomposition Applied mathematics PhD candidate, physics MA UC Berkeley April 26, 2013 UCB 1/19 Symmetric positive-definite I Definition A symmetric matrix A R n n is positive definite iff x T Ax > 0 holds x 0 R n.

More information

Lecture 8: Interest Point Detection. Saad J Bedros

Lecture 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 information

Finite Difference Methods (FDMs) 1

Finite Difference Methods (FDMs) 1 Finite Difference Methods (FDMs) 1 1 st - order Approxima9on Recall Taylor series expansion: Forward difference: Backward difference: Central difference: 2 nd - order Approxima9on Forward difference: Backward

More information

Advanced Machine Learning & Perception

Advanced Machine Learning & Perception Advanced Machine Learning & Perception Instructor: Tony Jebara Topic 1 Introduction, researchy course, latest papers Going beyond simple machine learning Perception, strange spaces, images, time, behavior

More information

Convex Optimization Algorithms for Machine Learning in 10 Slides

Convex Optimization Algorithms for Machine Learning in 10 Slides Convex Optimization Algorithms for Machine Learning in 10 Slides Presenter: Jul. 15. 2015 Outline 1 Quadratic Problem Linear System 2 Smooth Problem Newton-CG 3 Composite Problem Proximal-Newton-CD 4 Non-smooth,

More information

Inverse Problems, Information Content and Stratospheric Aerosols

Inverse Problems, Information Content and Stratospheric Aerosols Inverse Problems, Information Content and Stratospheric Aerosols May 22, 214 Table of contents 1 Inverse Problems 2 Aerosols 3 OSIRIS 4 Information Content 5 Non-Linear Techniques 6 Results Inverse Problems

More information

Programming, numerics and optimization

Programming, numerics and optimization Programming, numerics and optimization Lecture C-3: Unconstrained optimization II Łukasz Jankowski ljank@ippt.pan.pl Institute of Fundamental Technological Research Room 4.32, Phone +22.8261281 ext. 428

More information

Solving Linear Systems of Equations

Solving Linear Systems of Equations Solving Linear Systems of Equations Gerald Recktenwald Portland State University Mechanical Engineering Department gerry@me.pdx.edu These slides are a supplement to the book Numerical Methods with Matlab:

More information

Chapter 2: Matrix Algebra

Chapter 2: Matrix Algebra Chapter 2: Matrix Algebra (Last Updated: October 12, 2016) These notes are derived primarily from Linear Algebra and its applications by David Lay (4ed). Write A = 1. Matrix operations [a 1 a n. Then entry

More information

Multivariate Statistical Analysis

Multivariate Statistical Analysis Multivariate Statistical Analysis Fall 2011 C. L. Williams, Ph.D. Lecture 4 for Applied Multivariate Analysis Outline 1 Eigen values and eigen vectors Characteristic equation Some properties of eigendecompositions

More information

Introduction to Machine Learning

Introduction to Machine Learning 10-701 Introduction to Machine Learning PCA Slides based on 18-661 Fall 2018 PCA Raw data can be Complex, High-dimensional To understand a phenomenon we measure various related quantities If we knew what

More information

Numerical Linear Algebra. Computer Labs

Numerical Linear Algebra. Computer Labs Numerical Linear Algebra. Computer Labs 1 Notes To pass this course you should do any 2 computer assignments described below. You can work in groups by 2 persons. Sent final report for every computer assignment

More information

Learning strategies for neuronal nets - the backpropagation algorithm

Learning strategies for neuronal nets - the backpropagation algorithm Learning strategies for neuronal nets - the backpropagation algorithm In contrast to the NNs with thresholds we handled until now NNs are the NNs with non-linear activation functions f(x). The most common

More information

Optimization 2. CS5240 Theoretical Foundations in Multimedia. Leow Wee Kheng

Optimization 2. CS5240 Theoretical Foundations in Multimedia. Leow Wee Kheng Optimization 2 CS5240 Theoretical Foundations in Multimedia Leow Wee Kheng Department of Computer Science School of Computing National University of Singapore Leow Wee Kheng (NUS) Optimization 2 1 / 38

More information

Affine invariant Fourier descriptors

Affine invariant Fourier descriptors Affine invariant Fourier descriptors Sought: a generalization of the previously introduced similarityinvariant Fourier descriptors H. Burkhardt, Institut für Informatik, Universität Freiburg ME-II, Kap.

More information

Chapter 3 Numerical Methods

Chapter 3 Numerical Methods Chapter 3 Numerical Methods Part 2 3.2 Systems of Equations 3.3 Nonlinear and Constrained Optimization 1 Outline 3.2 Systems of Equations 3.3 Nonlinear and Constrained Optimization Summary 2 Outline 3.2

More information

Vasil Khalidov & Miles Hansard. C.M. Bishop s PRML: Chapter 5; Neural Networks

Vasil Khalidov & Miles Hansard. C.M. Bishop s PRML: Chapter 5; Neural Networks C.M. Bishop s PRML: Chapter 5; Neural Networks Introduction The aim is, as before, to find useful decompositions of the target variable; t(x) = y(x, w) + ɛ(x) (3.7) t(x n ) and x n are the observations,

More information

SUPPLEMENTARY INFORMATION

SUPPLEMENTARY INFORMATION scos SL PLB2 DOPC PL P BS PLB1 CL SF PL ND PLS O O BD PT IF DBS DD BS L3 CL SF PL ND BD BD f 0 OI L2 UST L1 S OC Quality assurance arm CCD f 0 + f US AO 50 Hz f 0 + f US AO 50 Hz PBS HWP BD RL RDD CL SF

More information