lecture 8: feature matching cs6670: computer vision noah snavely

16
Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Upload: arlene-florence-mcdaniel

Post on 18-Jan-2016

216 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Lecture 8: Feature matching

CS6670: Computer VisionNoah Snavely

Page 2: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

SIFT Example

siftsift

868 SIFT features

Page 3: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Feature matching

Given a feature in I1, how to find the best match in I2?1. Define distance function that compares two

descriptors2. Test all the features in I2, find the one with min

distance

Page 4: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Feature distanceHow to define the difference between two features f1, f2?

– Simple approach: L2 distance, ||f1 - f2 ||

– can give good scores to ambiguous (incorrect) matches

I1 I2

f1 f2

Page 5: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

f1 f2f2'

Feature distanceHow to define the difference between two features f1, f2?

• Better approach: ratio distance = ||f1 - f2 || / || f1 - f2’ || • f2 is best SSD match to f1 in I2

• f2’ is 2nd best SSD match to f1 in I2

• gives large values for ambiguous matches

I1 I2

Page 6: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Feature matching example

51 matches

Page 7: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Feature matching example

58 matches

Page 8: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Evaluating the resultsHow can we measure the performance of a feature matcher?

5075

200

feature distance

Page 9: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

True/false positives

The distance threshold affects performance– True positives = # of detected matches that are correct

• Suppose we want to maximize these—how to choose threshold?– False positives = # of detected matches that are incorrect

• Suppose we want to minimize these—how to choose threshold?

5075

200false match

true match

feature distance

How can we measure the performance of a feature matcher?

Page 10: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

0.7

Evaluating the results

0 1

1

false positive rate

truepositive

rate

# true positives# correctly matched features (positives)

0.1

How can we measure the performance of a feature matcher?

“recall”

# false positives# incorrectly matched features (negatives)

1 - “precision”

Page 11: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

0.7

Evaluating the results

0 1

1

false positive rate

truepositive

rate

0.1

ROC curve (“Receiver Operator Characteristic”)

How can we measure the performance of a feature matcher?

# true positives# correctly matched features (positives)

“recall”

# false positives# incorrectly matched features (negatives)

1 - “precision”

Page 12: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Lots of applications

Features are used for:– Image alignment (e.g., mosaics)– 3D reconstruction– Motion tracking– Object recognition (e.g., Google Goggles)– Indexing and database retrieval– Robot navigation– … other

Page 13: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Object recognition (David Lowe)

Page 14: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

3D Reconstruction

Internet Photos (“Colosseum”) Reconstructed 3D cameras and points

Page 15: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Sony Aibo

SIFT usage:

Recognize charging station

Communicate with visual cards

Teach object recognition

Page 16: Lecture 8: Feature matching CS6670: Computer Vision Noah Snavely

Questions?