CHAPTER 6

Question 1

Overview

Answer 1

• Changes are Everywhere!
• Why Estimate Motion ?
• Change Detection in a Video
• Two Approaches of Motion Tracking
• Feature-based
• Intensity gradient-based
• Lucas-Kanade Optical Flow Method
• Project

Question 2

Why Estimate Motion ?

Answer 2

• We live in a 4-D world
• Wide applications
• Object Tracking
• Camera Stabilization
• Image Mosaics
• 3D Shape Reconstruction
• Special Effects (Match Move)

Question 3

Any difference?

Answer 3

Yes,
position change

Question 4

Feature-based Motion Tracking

Answer 4

 Look for distinct features that change positions.

 Eagle’s wing tips change positions.
 Tree tops don’t change positions.

Question 5

Summary

Answer 5

 Simple algorithm.
 Can be slow if search area is large.
 Can constrain search area with prior knowledge.

Question 6

Stable (Good) Features

Answer 6

“flat” region:
no change in all
directions

“edge”:
no change along the
edge direction

“corner”:
significant change
in all directions
(uniqueness)

Question 7

slide 35

Answer 7

done

Question 8

 Lucas-Kanade algorithm is often used with Tomasi’s feature

Answer 8

 Apply Tomasi’s method to detect good features.
 Apply LK method to compute d for each pixel.
 Accept d only for good features.

Question 9

How to handle large displacement?

Answer 9

 What if we scale down images?
 Displacements are smaller!