DD2429 Computational Photography :00-19:00

Similar documents
ECE 470: Homework 5. Due Tuesday, October 27 in Seth Hutchinson. Luke A. Wendt

Introduction to Homogeneous coordinates

Computer Vision. Coordinates. Prof. Flávio Cardeal DECOM / CEFET- MG.

Revision Problems for Examination 2 in Algebra 1

TD2 : Stereoscopy and Tracking: solutions

Computer Graphics 7: Viewing in 3-D

Geometric Computing. in Image Analysis and Visualization. Lecture Notes March 2007

CS223b Midterm Exam, Computer Vision. Monday February 25th, Winter 2008, Prof. Jana Kosecka

E0005E - Industrial Image Analysis

Short on camera geometry and camera calibration

Multiple View Geometry in Computer Vision

CS6670: Computer Vision

1 Projective Geometry

Camera Model and Calibration

COMP 558 lecture 19 Nov. 17, 2010

7. The Gauss-Bonnet theorem

calibrated coordinates Linear transformation pixel coordinates

Robotics - Projective Geometry and Camera model. Marcello Restelli

3D Modeling using multiple images Exam January 2008

CIS 580, Machine Perception, Spring 2016 Homework 2 Due: :59AM

Visual Recognition: Image Formation

Chapter 8 Three-Dimensional Viewing Operations

Math 26: Fall (part 1) The Unit Circle: Cosine and Sine (Evaluating Cosine and Sine, and The Pythagorean Identity)

Geometric camera models and calibration

More on single-view geometry class 10

WHAT YOU SHOULD LEARN

Camera Models and Image Formation. Srikumar Ramalingam School of Computing University of Utah

Math 231E, Lecture 34. Polar Coordinates and Polar Parametric Equations

Answers to practice questions for Midterm 1

Chapter 6. Curves and Surfaces. 6.1 Graphs as Surfaces

Camera Models and Image Formation. Srikumar Ramalingam School of Computing University of Utah

Computer Vision: Lecture 3

Geometric Algebra. 8. Conformal Geometric Algebra. Dr Chris Doran ARM Research

A GENTLE INTRODUCTION TO THE BASIC CONCEPTS OF SHAPE SPACE AND SHAPE STATISTICS

Intersection of an Oriented Box and a Cone

Differential Geometry: Circle Patterns (Part 1) [Discrete Conformal Mappinngs via Circle Patterns. Kharevych, Springborn and Schröder]

1 Affine and Projective Coordinate Notation

Mathematics of a Multiple Omni-Directional System

Computer Graphics: Geometric Transformations

CIS 580, Machine Perception, Spring 2015 Homework 1 Due: :59AM

3D Sensing. 3D Shape from X. Perspective Geometry. Camera Model. Camera Calibration. General Stereo Triangulation.

Calculus II. Step 1 First, here is a quick sketch of the graph of the region we are interested in.

A Constant Rate of Change Name Part 1

Pin Hole Cameras & Warp Functions

The diagram above shows a sketch of the curve C with parametric equations

Math for Geometric Optics

Camera Model and Calibration. Lecture-12

Hello, welcome to the video lecture series on Digital Image Processing. So in today's lecture

Polar Coordinates. Chapter 10: Parametric Equations and Polar coordinates, Section 10.3: Polar coordinates 27 / 45

COSC579: Scene Geometry. Jeremy Bolton, PhD Assistant Teaching Professor

COMP30019 Graphics and Interaction Three-dimensional transformation geometry and perspective

Visualizing Quaternions

Rectangular Coordinates in Space

Figure 1. Lecture 1: Three Dimensional graphics: Projections and Transformations

Camera Calibration and Light Source Estimation from Images with Shadows

Computer Graphics: Viewing in 3-D. Course Website:

To Do. Outline. Translation. Homogeneous Coordinates. Foundations of Computer Graphics. Representation of Points (4-Vectors) Start doing HW 1

MATHEMATICS FOR ENGINEERING TUTORIAL 5 COORDINATE SYSTEMS

Suggested problems - solutions

Kevin James. MTHSC 206 Section 15.6 Directional Derivatives and the Gra

Advanced Operations Research Techniques IE316. Quiz 1 Review. Dr. Ted Ralphs

Contents. 1 Introduction Background Organization Features... 7

Single-view 3D Reconstruction

Perspective projection and Transformations

Camera Calibration. Schedule. Jesus J Caban. Note: You have until next Monday to let me know. ! Today:! Camera calibration

PreCalculus Unit 1: Unit Circle Trig Quiz Review (Day 9)

N-Views (1) Homographies and Projection

CSE 252B: Computer Vision II

Multiple View Geometry in computer vision

Viewing with Computers (OpenGL)

Midterm Exam Fundamentals of Computer Graphics (COMP 557) Thurs. Feb. 19, 2015 Professor Michael Langer

Ch. 2 Trigonometry Notes

Lesson 5.6: Angles in Standard Position

3D Geometry and Camera Calibration

Final Examination. Math1339 (C) Calculus and Vectors. December 22, :30-12:30. Sanghoon Baek. Department of Mathematics and Statistics

Computer Vision I - Algorithms and Applications: Multi-View 3D reconstruction

Viewing. Reading: Angel Ch.5

AQA GCSE Further Maths Topic Areas

CSE 252B: Computer Vision II

Problem Possible Points Points Earned Problem Possible Points Points Earned Test Total 100

EE 584 MACHINE VISION

Camera Projection Models We will introduce different camera projection models that relate the location of an image point to the coordinates of the

Multiple Views Geometry

Vector Addition. Qty Item Part Number 1 Force Table ME-9447B 1 Mass and Hanger Set ME Carpenter s level 1 String

2D Object Definition (1/3)

A simple method for interactive 3D reconstruction and camera calibration from a single view

ME 115(b): Final Exam, Spring

Parallel and perspective projections such as used in representing 3d images.

Lecture 3: Camera Calibration, DLT, SVD

Introduction to Computer Vision

CSC 305 The Graphics Pipeline-1

Rotations (and other transformations) Rotation as rotation matrix. Storage. Apply to vector matrix vector multiply (15 flops)

Today s lecture. Image Alignment and Stitching. Readings. Motion models

Pin Hole Cameras & Warp Functions

Generalized barycentric coordinates

Computer Vision Project-1

Math (Spring 2009): Lecture 5 Planes. Parametric equations of curves and lines

Translations. Geometric Image Transformations. Two-Dimensional Geometric Transforms. Groups and Composition

A Geometric Analysis of Subspace Clustering with Outliers

10.1 Curves Defined by Parametric Equations

Transcription:

. Examination: DD2429 Computational Photography 202-0-8 4:00-9:00 Each problem gives max 5 points. In order to pass you need about 0-5 points. You are allowed to use the lecture notes and standard list of mathematical formulas. A perfect sphere is viewed by a normalised camera x y = λ a 0 0 0 0 0 0 0 0 0 the -axis intersects the center of the sphere and the original image is therefore a perfect circle with center on the origin of the image (x, y) coordinate system If we change the camera parameters the projection is described by: x y = λ σ x γ x 0 0 σ y y 0 0 0 r r 2 r 3 r 2 r 22 r 23 r 3 r 32 r 33 0 0 0 0 0 Y 0 0 0 0 The figure shows the image of the sphere before and after a change of the camera parameters Denote for each of the changes in camera parameters a-e below the associated images -5. One image for each parameter change. Different parameter alternatives could map to the same image a) Move the camera position 0 forwards along the -axis b) Change of the principal point c) Rotate the camera around the Y-axis d) Move the camera position 0 backwards along the -axis and change the focal length f so that the image plane remains in the same position e) Change of σ x Whatever looks like a perfect circle (images, 2, 4, 5) should be considered as a perfect circle. 2. A normalized camera is located at (t, t Y, t ) The camera can be expressed as: x y = λ 0 0 t 0 0 t Y 0 0 t

Figure : 2

a) Formulate a set of linear equations for finding the camera location (t, t Y, t ) assuming that you know the 3D and image coordinates of a set of points. b Show that this linear system of three unknowns can be separated into two systems of two unknowns (t, t ) and (t Y, t ) c) Solve one of this systems using a minimal set of points. I.e give an explicit solution of (t, t ) in terms of image and 3D coordinates 3 A general uncalibrated perspective camera is looking at a planar surface where there is a unit square. Any point on the plane can then be expressed in the coordinate system (x, y ) defined by the unit square where one of the corners is used as the origin. a) What is the general exact relation between the coordinates (x, y ) of a point on the plane and it s image coordinates (x, y)? (No derivations, you only have to state the result) b) Describe in words how you would go about computing the coordinates (x, y ) of a point on the plane from knowledge of it s image coordinates, using the image coordinates of the corners of the square Again you don t have to compute anything just formulate the equations, what operations have to be performed in order to solve them, what information you use and in what order 4 Figure 2: We have two cameras: a) A normalized camera located at the origin and b) the same camera rotated the angle θ around the axis and moved distance t along the axis xa y a = λ 0 0 0 0 0 0 0 0 0 cos(θ) sin(θ) 0 y b = λ sin(θ) cos(θ) 0 0 0 xb 0 0 0 0 0 0 0 0 t 3

Suppose the cameras are looking at a unit circle 2 + Y 2 = located in a plane orthogonal to the axis at distance d from the origin. a) Compute the images of this unit circle in camera a and b b) Can the translation t and rotation θ of the second camera be determined from knowledge of the images of the unit circle in the two cameras? Give a motivation from the equations of the circles in the two images as well as a geometric motivation 5 A line in 3-D space passes through two points P = (, Y, ) T and P 2 = ( 2, Y 2, 2 ) T. All points on the line can then be expressed as with P = (, Y,, ) T where µ is a real number P = P + µ(p 2 P ) A camera with general 3 4 matrix M is looking at the line a) Assume that the line in space is infinitely long i.e. < µ < + and that only the part of the line in front of the camera will be projected The projection of the line in the image will then be of finite extent Sketch a geometric explanation for this. (Use the figure in the problem) b) If we assume that the line is in front of the camera for all µ > 0 then the point on the line corresponding to lim µ will get projected to a specific point in the image. Compute the image coordinates of this point in terms of the camera matrix M and the points P and P 2 c) If the line in space was moved parallel to it s original position What would be the respective solutions to a) and b)? Hint: The crucial problem is to understand how λ(µ) varies with µ You can assume a specific scale factor for M and use the form: p = λ(µ)m P Note that the real number λ(µ) is only there to ensure that the image coordinate p = (x, y, ) 4

Figure 3: 5