Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 19 Motion.

Slides:



Advertisements
Similar presentations
1 Copyright © 2010, Elsevier Inc. All rights Reserved Fig 4.1 Chapter 4.
Advertisements

1 Copyright © 2010, Elsevier Inc. All rights Reserved Fig 6.1 Chapter 6.
Dynamic Occlusion Analysis in Optical Flow Fields
© 2012 Pearson Education, Inc. { Chapter 27 Magnetic Fields and Forces (cont.)
Computer Vision Optical Flow
Optical Flow Methods 2007/8/9.
Visual motion Many slides adapted from S. Seitz, R. Szeliski, M. Pollefeys.
3D Rigid/Nonrigid RegistrationRegistration 1)Known features, correspondences, transformation model – feature basedfeature based 2)Specific motion type,
Computer Vision Spring ,-685 Instructor: S. Narasimhan Wean 5403 T-R 3:00pm – 4:20pm Lecture #15.
Optical flow (motion vector) computation Course: Computer Graphics and Image Processing Semester:Fall 2002 Presenter:Nilesh Ghubade
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”
Curl and Divergence.
December 9, 2014Computer Vision Lecture 23: Motion Analysis 1 Now we will talk about… Motion Analysis.
Course 11 Optical Flow. 1. Concept Observe the scene by moving viewer Optical flow provides a clue to recover the motion. 2. Constraint equation.
Motion Analysis using Optical flow CIS750 Presentation Student: Wan Wang Prof: Longin Jan Latecki Spring 2003 CIS Dept of Temple.
3D Imaging Motion.
1 Motion Analysis using Optical flow CIS601 Longin Jan Latecki Fall 2003 CIS Dept of Temple University.
Dynamic Perspective How a moving camera reveals scene depth and egomotion parameters Jitendra Malik UC Berkeley.
Solar Magnetic Fields. Capacitors in Circuits Charge takes time to move through wire  V is felt at the speed of light, however Change in potential across.
CSSE463: Image Recognition Day 29
Tangent Vectors and Normal Vectors
Presented by: Cindy Yan EE6358 Computer Vision
Modeling Constraints with Parametrics
CSSE463: Image Recognition Day 29
Copyright © 2016 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
Chapter 11.
CSSE463: Image Recognition Day 29
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
Chapter 10.
Chapter 28 - Renal Hyperplasia and Hypertrophy
CSSE463: Image Recognition Day 29
Copyright © 2012, Elsevier Inc. All rights Reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Modeling Cross-Cutting Relationships with Allocations
Portable Biotechnology
© 2012 Elsevier, Inc. All rights reserved.
CSSE463: Image Recognition Day 29
Modeling Text-Based Requirements and their Relationship to Design
Modeling Functionality with Use Cases
Customizing SysML for Specific Domains
Copyright © 2012, Elsevier Inc. All rights Reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 12.
Chapter 6.
Copyright © 2012, Elsevier Inc. All rights Reserved.
© 2012 Elsevier, Inc. All rights reserved.
Chapter 103 Long-Term Care: The Global Impact
Chapter 01.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 08.
Chapter 64 - Renal Calcium Metabolism
© 2015 Elsevier, Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2013 Elsevier Inc. All rights reserved.
Copyright © 2012, Elsevier Inc. All rights Reserved.
Chapter 15 Contraception
Copyright © 2013 Elsevier Inc. All rights reserved.
Chapter 15.
Chapter 77 Cervical Cancer: Burden of Disease
Chapter 20 Assisted Reproductive Technologies
Chapter 3.
© 2015 Elsevier, Inc. All rights reserved.
Presentation transcript:

Copyright © 2012 Elsevier Inc. All rights reserved.. Chapter 19 Motion

19.2 Optical Flow When scenes contain moving objects, analysis is necessarily more complex than for scenes where every is stationary. Image differencing over successive pairs of frames is a straightforward method. More careful consideration shows that things are not quite so simple (Fig. 19.1) The edges parallel to the motion direction do not show up. Regions of constant intensity give no sign of motion. Copyright © 2012 Elsevier Inc. All rights reserved.. 2

FIGURE 19.1 Copyright © 2012 Elsevier Inc. All rights reserved.. 3

Optical Flow: a local operator is applied at all pixels in the image and lead to a motion vector field that varies smoothly over the whole image. The attraction lies in the use of a local operator, with its limited computational burden. We start by considering the intensity function I(x,y,t) and expand it in a Taylor series: Copyright © 2012 Elsevier Inc. All rights reserved.. 4

Ignoring the second and higher order items, I x, I y and I t denote respective partial derivatives with respect to x, y, and t. Next, we set the local condition that the image has shifted by amount (dx, dy) in time dt so that Hence, we can deduce Copyright © 2012 Elsevier Inc. All rights reserved.. 5

FIGURE 19.2 Copyright © 2012 Elsevier Inc. All rights reserved.. 6

Let the local velocity v in the form: We find: The component of v lie on the following line: Therefore the normal to the direction (I x, I y ): Copyright © 2012 Elsevier Inc. All rights reserved.. 7

FIGURE 19.3 Copyright © 2012 Elsevier Inc. All rights reserved.. 8

19.3 Interpretation of Optical Flow Fields Copyright © 2012 Elsevier Inc. All rights reserved.. 9 FIGURE 19.4 Interpretation of velocity flow fields. (a) A case where the object features all have zero velocity. (b) A case where an object is moving to the right. (c) A case where the camera is moving into the scene, and the stationary object features appear to be diverging from a focus of expansion (FOE), while a single large object is moving past the camera and away from a separate FOE. In (d), an object is moving directly toward the camera that is stationary: the object’s FOE lies within its outline. In (e), an object is rotating about the line of sight to the camera, and in (f), the object is rotating about an axis perpendicular to the line of sight. In all cases, the length of the arrow indicates the magnitude of the velocity vector