Dynamic Perspective How a moving camera reveals scene depth and egomotion parameters Jitendra Malik UC Berkeley.

Slides:



Advertisements
Similar presentations
Epipolar Geometry.
Advertisements

Computer vision. Camera Calibration Camera Calibration ToolBox – Intrinsic parameters Focal length: The focal length in pixels is stored in the.
Camera Models A camera is a mapping between the 3D world and a 2D image The principal camera of interest is central projection.
Announcements. Projection Today’s Readings Nalwa 2.1.
CS485/685 Computer Vision Prof. George Bebis
CSc83029 – 3-D Computer Vision/ Ioannis Stamos 3-D Computational Vision CSc Optical Flow & Motion The Factorization Method.
Computer Vision : CISC4/689
© 2003 by Davi GeigerComputer Vision October 2003 L1.1 Structure-from-EgoMotion (based on notes from David Jacobs, CS-Maryland) Determining the 3-D structure.
Multiple View Geometry Marc Pollefeys University of North Carolina at Chapel Hill Modified by Philippos Mordohai.
Single-view geometry Odilon Redon, Cyclops, 1914.
The Pinhole Camera Model
Projected image of a cube. Classical Calibration.
Motion Field and Optical Flow. Outline Motion Field and Optical Flow Definition, Example, Relation Optical Flow Constraint Equation Assumptions & Derivation,
CS223b, Jana Kosecka Rigid Body Motion and Image Formation.
3D Motion Estimation. 3D model construction Video Manipulation.
Camera parameters Extrinisic parameters define location and orientation of camera reference frame with respect to world frame Intrinsic parameters define.
Motion from normal flow. Optical flow difficulties The aperture problemDepth discontinuities.
Stockman MSU/CSE Math models 3D to 2D Affine transformations in 3D; Projections 3D to 2D; Derivation of camera matrix form.
3D Computer Vision and Video Computing 3D Vision Topic 8 of Part 2 Visual Motion (II) CSC I6716 Spring 2004 Zhigang Zhu, NAC 8/203A
Computer Vision Spring ,-685 Instructor: S. Narasimhan Wean 5403 T-R 3:00pm – 4:20pm Lecture #15.
3-D Scene u u’u’ Study the mathematical relations between corresponding image points. “Corresponding” means originated from the same 3D point. Objective.
A plane-plus-parallax algorithm Basic Model: When FOV is not very large and the camera motion has a small rotation, the 2D displacement (u,v) of an image.
Viewing and Projections
Course 12 Calibration. 1.Introduction In theoretic discussions, we have assumed: Camera is located at the origin of coordinate system of scene.
CSE 681 Review: Transformations. CSE 681 Transformations Modeling transformations build complex models by positioning (transforming) simple components.
Imaging Geometry for the Pinhole Camera Outline: Motivation |The pinhole camera.
1-1 Measuring image motion velocity field “local” motion detectors only measure component of motion perpendicular to moving edge “aperture problem” 2D.
1 Computational Vision CSCI 363, Fall 2012 Lecture 31 Heading Models.
Computer Vision, Robert Pless Lecture 11 our goal is to understand the process of multi-camera vision. Last time, we studies the “Essential” and “Fundamental”
视觉的三维运动理解 刘允才 上海交通大学 2002 年 11 月 16 日 Understanding 3D Motion from Images Yuncai Liu Shanghai Jiao Tong University November 16, 2002.
Metrology 1.Perspective distortion. 2.Depth is lost.
Correspondence-Free Determination of the Affine Fundamental Matrix (Tue) Young Ki Baik, Computer Vision Lab.
Vision Review: Image Formation Course web page: September 10, 2002.
Course 11 Optical Flow. 1. Concept Observe the scene by moving viewer Optical flow provides a clue to recover the motion. 2. Constraint equation.
Viewing CS418 Computer Graphics John C. Hart. Graphics Pipeline Homogeneous Divide Model Coords Model Xform World Coords Viewing Xform Still Clip Coords.
Single-view geometry Odilon Redon, Cyclops, 1914.
CS-498 Computer Vision Week 7, Day 2 Camera Parameters Intrinsic Calibration  Linear  Radial Distortion (Extrinsic Calibration?) 1.
3D Imaging Motion.
CS332 Visual Processing Department of Computer Science Wellesley College Analysis of Motion Recovering observer motion.
Assigned work: pg.398 #1,3,7,8,11-14 pg. 419 #3 (work) Recall: Dot Product.
1 Motion Analysis using Optical flow CIS601 Longin Jan Latecki Fall 2003 CIS Dept of Temple University.
Computer Graphics 3D Transformations. Translation.
CS332 Visual Processing Department of Computer Science Wellesley College Analysis of Motion Recovering 3-D structure from motion.
12/24/2015 A.Aruna/Assistant professor/IT/SNSCE 1.
Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 19 Motion.
Multiple Light Source Optical Flow Multiple Light Source Optical Flow Robert J. Woodham ICCV’90.
Course14 Dynamic Vision. Biological vision can cope with changing world Moving and changing objects Change illumination Change View-point.
True FOE Computed FOE Recovering the observer’s rotation Velocity component due to observer’s translation Velocity component due to observer’s rotation.
Computer vision: models, learning and inference M Ahad Multiple Cameras
Lecture 9 Feature Extraction and Motion Estimation Slides by: Michael Black Clark F. Olson Jean Ponce.
Determining 3D Structure and Motion of Man-made Objects from Corners.
Digital Image Processing Additional Material : Imaging Geometry 11 September 2006 Digital Image Processing Additional Material : Imaging Geometry 11 September.
INTRODUCTION TO DYNAMICS ANALYSIS OF ROBOTS (Part 1)
CS 551 / 645: Introductory Computer Graphics Viewing Transforms.
Transformations University of British Columbia CPSC 314 Computer Graphics Jan-Apr 2016 Tamara Munzner.
CS682, Jana Kosecka Rigid Body Motion and Image Formation Jana Kosecka
55:148 Digital Image Processing Chapter 11 3D Vision, Geometry
Visual motion Many slides adapted from S. Seitz, R. Szeliski, M. Pollefeys.
Chapter 9 Vector Calculus.
Modeling 101 For the moment assume that all geometry consists of points, lines and faces Line: A segment between two endpoints Face: A planar area bounded.
3D Motion Estimation.
Planar Kinematics of a Rigid Body: Review II
Two-view geometry.
Conceptual Dynamics Part II: Kinematics of Particles Chapter 3
Viewing and Perspective Transformations
Multiple View Geometry for Robotics
Course 7 Motion.
Chapter 3 Vectors Questions 3-1 Vectors and Scalars
The Pinhole Camera Model
Presentation transcript:

Dynamic Perspective How a moving camera reveals scene depth and egomotion parameters Jitendra Malik UC Berkeley

The Pinhole Camera x y

Suppose the camera moves with respect to the world… are the two components of the optical flow at (x,y)

Gibson’s example I: Optical flow for a pilot landing a plane

Gibson’s example II: Optical flow from the side window of a car

Outline Derive equation relating optical flow field to scene depth Z(x,y) and the motion of the camera t, ω The translational component of the flow field is the more important one – it is what tells Z(x,y) and the translation t The rotational component of the flow field reveals information about ω

How does a point X in the scene move?

Now consider the effect of projection…

The Optical Flow Equations

Optical flow for pure translation

Optical flow for pure translation along Z axis

Optical flow for pure translation along Z-axis The optical flow vector is a scalar multiple of the position vector

Scale Factor Ambiguity However we can compute the time to contact

Optical flow for general translation When is this (0,0)?

With respect to the FOE, the flow vectors are radially outward

The Optical Flow Equations