![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
|
|
|
|
|
|
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
Suppose we have
more than two images
|
|
|
|
• |
How to track a
point through all of the images?
|
|
|
|
![](v3_slide0114_image048.gif) |
|
|
|
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
![](v3_space.gif) |
Feature Tracking
|
|
|
|
• |
Choose only the
points (“features”) that are easily tracked
|
|
|
• |
How to find these
features?
|
|
|
|
|
|
![](v3_space.gif) |
![](v3_space.gif) |
– |
windows
where has
two large eigenvalues
|
|
|
|
|
![](v3_space.gif) |
![](v3_space.gif) |
• |
Called the Harris
Corner Detector
|
|
|
|