CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.
-
Upload
myron-chase -
Category
Documents
-
view
218 -
download
0
Transcript of CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.
![Page 1: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/1.jpg)
CS 558 COMPUTER VISIONLecture VII: Corner and Blob Detection
Slides adapted from S. Lazebnik
![Page 2: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/2.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 3: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/3.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 4: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/4.jpg)
FEATURE EXTRACTION: CORNERS9300 Harris Corners Pkwy, Charlotte, NC
![Page 5: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/5.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 6: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/6.jpg)
WHY EXTRACT FEATURES?
• Motivation: panorama stitching We have two images – how do we combine
them?
![Page 7: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/7.jpg)
WHY EXTRACT FEATURES?
• Motivation: panorama stitching We have two images – how do we combine
them?
Step 1: extract featuresStep 2: match features
![Page 8: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/8.jpg)
WHY EXTRACT FEATURES?
• Motivation: panorama stitching We have two images – how do we combine
them?
Step 1: extract featuresStep 2: match featuresStep 3: align images
![Page 9: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/9.jpg)
CHARACTERISTICS OF GOOD FEATURES
• Repeatability The same feature can be found in several images despite
geometric and photometric transformations • Saliency
Each feature is distinctive• Compactness and efficiency
Many fewer features than image pixels• Locality
A feature occupies a relatively small area of the image; robust to clutter and occlusion
![Page 10: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/10.jpg)
APPLICATIONS
Feature points are used for: Image alignment 3D reconstructionMotion trackingRobot navigation Indexing and database retrievalObject recognition
![Page 11: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/11.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 12: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/12.jpg)
FINDING CORNERS
• Key property: in the region around a corner, image gradient has two or more dominant directions
• Corners are repeatable and distinctive
C.Harris and M.Stephens. "A Combined Corner and Edge Detector.“ Proceedings of the 4th Alvey Vision Conference: pages 147—151, 1988.
![Page 13: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/13.jpg)
CORNER DETECTION: BASIC IDEA
• We should easily recognize the point by looking through a small window
• Shifting a window in any direction should give a large change in intensity
“edge”:no change along the edge direction
“corner”:significant change in all directions
“flat” region:no change in all directions
Source: A. Efros
![Page 14: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/14.jpg)
CORNER DETECTION: MATHEMATICS
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y
Change in appearance of window w(x,y) for the shift [u,v]:
I(x, y)E(u, v)
E(3,2)
w(x, y)
![Page 15: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/15.jpg)
CORNER DETECTION: MATHEMATICS
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y
I(x, y)E(u, v)
E(0,0)
w(x, y)
Change in appearance of window w(x,y) for the shift [u,v]:
![Page 16: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/16.jpg)
CORNER DETECTION: MATHEMATICS
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y
IntensityShifted intensity
Window function
orWindow function w(x,y) =
Gaussian1 in window, 0 outside
Source: R. Szeliski
Change in appearance of window w(x,y) for the shift [u,v]:
![Page 17: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/17.jpg)
CORNER DETECTION: MATHEMATICS
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y
We want to find out how this function behaves for small shifts
Change in appearance of window w(x,y) for the shift [u,v]:
E(u, v)
![Page 18: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/18.jpg)
CORNER DETECTION: MATHEMATICS
v
u
EE
EEvu
E
EvuEvuE
vvuv
uvuu
v
u
)0,0()0,0(
)0,0()0,0(][
2
1
)0,0(
)0,0(][)0,0(),(
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y
Local quadratic approximation of E(u,v) in the neighborhood of (0,0) is given by the second-order Taylor expansion:
We want to find out how this function behaves for small shifts
Change in appearance of window w(x,y) for the shift [u,v]:
![Page 19: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/19.jpg)
CORNER DETECTION: MATHEMATICS
v
u
EE
EEvu
E
EvuEvuE
vvuv
uvuu
v
u
)0,0()0,0(
)0,0()0,0(][
2
1
)0,0(
)0,0(][)0,0(),(
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y Second-order Taylor expansion of E(u,v) about (0,0):
),(),(),(),(2
),(),(),(2),(
),(),(),(),(2
),(),(),(2),(
),(),(),(),(2),(
,
,
,
,
,
vyuxIyxIvyuxIyxw
vyuxIvyuxIyxwvuE
vyuxIyxIvyuxIyxw
vyuxIvyuxIyxwvuE
vyuxIyxIvyuxIyxwvuE
xyyx
xyyx
uv
xxyx
xxyx
uu
xyx
u
![Page 20: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/20.jpg)
CORNER DETECTION: MATHEMATICS
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y Second-order Taylor expansion of E(u,v) about (0,0):
),(),(),(2)0,0(
),(),(),(2)0,0(
),(),(),(2)0,0(
0)0,0(
0)0,0(
0)0,0(
,
,
,
yxIyxIyxwE
yxIyxIyxwE
yxIyxIyxwE
E
E
E
yxyx
uv
yyyx
vv
xxyx
uu
v
u
v
u
EE
EEvu
E
EvuEvuE
vvuv
uvuu
v
u
)0,0()0,0(
)0,0()0,0(][
2
1
)0,0(
)0,0(][)0,0(),(
![Page 21: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/21.jpg)
CORNER DETECTION: MATHEMATICS
v
u
yxIyxwyxIyxIyxw
yxIyxIyxwyxIyxw
vuvuE
yxy
yxyx
yxyx
yxx
,
2
,
,,
2
),(),(),(),(),(
),(),(),(),(),(
][),(
2
,
( , ) ( , ) ( , ) ( , )x y
E u v w x y I x u y v I x y Second-order Taylor expansion of E(u,v) about (0,0):
),(),(),(2)0,0(
),(),(),(2)0,0(
),(),(),(2)0,0(
0)0,0(
0)0,0(
0)0,0(
,
,
,
yxIyxIyxwE
yxIyxIyxwE
yxIyxIyxwE
E
E
E
yxyx
uv
yyyx
vv
xxyx
uu
v
u
![Page 22: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/22.jpg)
CORNER DETECTION: MATHEMATICS
v
uMvuvuE ][),(
The quadratic approximation simplifies to
2
2,
( , ) x x y
x y x y y
I I IM w x y
I I I
where M is a second moment matrix computed from image derivatives:
M
![Page 23: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/23.jpg)
The surface E(u,v) is locally approximated by a quadratic form. Let’s try to understand its shape.
INTERPRETING THE SECOND MOMENT MATRIX
v
uMvuvuE ][),(
yx yyx
yxx
III
IIIyxwM
,2
2
),(
![Page 24: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/24.jpg)
2
1
,2
2
0
0),(
yx yyx
yxx
III
IIIyxwM
First, consider the axis-aligned case (gradients are either horizontal or vertical)
If either λ is close to 0, then this is not a corner, so look for locations where both are large.
INTERPRETING THE SECOND MOMENT MATRIX
![Page 25: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/25.jpg)
Consider a horizontal “slice” of E(u, v):
INTERPRETING THE SECOND MOMENT MATRIX
const][
v
uMvu
This is the equation of an ellipse.
![Page 26: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/26.jpg)
Consider a horizontal “slice” of E(u, v):
INTERPRETING THE SECOND MOMENT MATRIX
const][
v
uMvu
This is the equation of an ellipse.
RRM
2
11
0
0
The axis lengths of the ellipse are determined by the eigenvalues and the orientation is determined by R
direction of the slowest change
direction of the fastest change
(max)-1/2
(min)-1/2
Diagonalization of M:
![Page 27: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/27.jpg)
VISUALIZATION OF SECOND MOMENT MATRICES
![Page 28: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/28.jpg)
VISUALIZATION OF SECOND MOMENT MATRICES
![Page 29: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/29.jpg)
INTERPRETING THE EIGENVALUES
1
2
“Corner”1 and 2 are large,
1 ~ 2;
E increases in all directions
1 and 2 are small;
E is almost constant in all directions
“Edge” 1 >> 2
“Edge” 2 >> 1
“Flat” region
Classification of image points using eigenvalues of M:
![Page 30: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/30.jpg)
CORNER RESPONSE FUNCTION
“Corner”R > 0
“Edge” R < 0
“Edge” R < 0
“Flat” region
|R| small
22121
2 )()(trace)det( MMR
α: constant (0.04 to 0.06)
![Page 31: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/31.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 32: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/32.jpg)
HARRIS DETECTOR: STEPS
1. Compute Gaussian derivatives at each pixel2. Compute second moment matrix M in a Gaussian
window around each pixel 3. Compute corner response function R4. Threshold R5. Find local maxima of response function
(nonmaximum suppression)
C.Harris and M.Stephens. “A Combined Corner and Edge Detector.” Proceedings of the 4th Alvey Vision Conference: pages 147—151, 1988.
![Page 33: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/33.jpg)
HARRIS DETECTOR: STEPS
![Page 34: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/34.jpg)
HARRIS DETECTOR: STEPSCompute corner response R
![Page 35: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/35.jpg)
HARRIS DETECTOR: STEPS
Find points with large corner response: R>threshold
![Page 36: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/36.jpg)
HARRIS DETECTOR: STEPS
Take only the points of local maxima of R
![Page 37: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/37.jpg)
HARRIS DETECTOR: STEPS
![Page 38: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/38.jpg)
INVARIANCE AND COVARIANCE
• We want corner locations to be invariant to photometric transformations and covariant to geometric transformations Invariance: image is transformed and corner locations do not
change Covariance: if we have two transformed versions of the same
image, features should be detected in corresponding locations
![Page 39: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/39.jpg)
AFFINE INTENSITY CHANGE
• Only derivatives are used => invariance to intensity shift I I + b
• Intensity scaling: I a I
R
x (image coordinate)
threshold
R
x (image coordinate)
Partially invariant to affine intensity change
I a I + b
![Page 40: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/40.jpg)
IMAGE TRANSLATION
• Derivatives and window function are shift-invariant
Corner location is covariant w.r.t. translation
![Page 41: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/41.jpg)
IMAGE ROTATION
Second moment ellipse rotates but its shape (i.e. eigenvalues) remains the same
Corner location is covariant w.r.t. rotation
![Page 42: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/42.jpg)
SCALING
All points will be classified as edges
Corner
Corner location is not covariant to scaling!
![Page 43: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/43.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 44: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/44.jpg)
BLOB DETECTION
![Page 45: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/45.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 46: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/46.jpg)
• Goal: independently detect corresponding regions in scaled versions of the same image
• Need scale selection mechanism for finding characteristic region size that is covariant with the image transformation
ACHIEVING SCALE COVARIANCE
![Page 47: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/47.jpg)
RECALL: EDGE DETECTION
gdx
df
f
gdx
d
Source: S. Seitz
Edge
Derivativeof Gaussian
Edge = maximumof derivative
![Page 48: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/48.jpg)
EDGE DETECTION, TAKE 2
gdx
df
2
2
f
gdx
d2
2
Edge
Second derivativeof Gaussian (Laplacian)
Edge = zero crossingof second derivative
Source: S. Seitz
![Page 49: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/49.jpg)
FROM EDGES TO BLOBS• Edge = ripple• Blob = superposition of two ripples
Spatial selection: the magnitude of the Laplacianresponse will achieve a maximum at the center ofthe blob, provided the scale of the Laplacian is“matched” to the scale of the blob
maximum
![Page 50: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/50.jpg)
• We want to find the characteristic scale of the blob by convolving it with Laplacians at several scales and looking for the maximum response
• However, Laplacian response decays as scale increases:
SCALE SELECTION
Why does this happen?
increasing σoriginal signal(radius=8)
![Page 51: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/51.jpg)
SCALE NORMALIZATION
• The response of a derivative of Gaussian filter to a perfect step edge decreases as σ increases
2
1
![Page 52: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/52.jpg)
SCALE NORMALIZATION
• The response of a derivative of Gaussian filter to a perfect step edge decreases as σ increases
• To keep response the same (scale-invariant), must multiply Gaussian derivative by σ
• Laplacian is the second Gaussian derivative, so it must be multiplied by σ2
![Page 53: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/53.jpg)
EFFECT OF SCALE NORMALIZATION
Scale-normalized Laplacian response
Unnormalized Laplacian responseOriginal signal
maximum
![Page 54: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/54.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 55: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/55.jpg)
BLOB DETECTION IN 2D
Laplacian of Gaussian: Circularly symmetric operator for blob detection in 2D
2
2
2
22
y
g
x
gg
![Page 56: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/56.jpg)
BLOB DETECTION IN 2D
Laplacian of Gaussian: Circularly symmetric operator for blob detection in 2D
2
2
2
222
norm y
g
x
gg Scale-normalized:
![Page 57: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/57.jpg)
• At what scale does the Laplacian achieve a maximum response to a binary circle of radius r?
SCALE SELECTION
r
image Laplacian
![Page 58: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/58.jpg)
• At what scale does the Laplacian achieve a maximum response to a binary circle of radius r?
• To get maximum response, the zeros of the Laplacian have to be aligned with the circle
• The Laplacian is given by:
• Therefore, the maximum response occurs at
SCALE SELECTION
r
image
62/)(222 2/)2(222
yxeyx .2/r
circle
Laplacian
0
![Page 59: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/59.jpg)
CHARACTERISTIC SCALE
• We define the characteristic scale of a blob as the scale that produces peak of Laplacian response in the blob center
characteristic scale
T. Lindeberg (1998). "Feature detection with automatic scale selection." International Journal of Computer Vision 30 (2): pp 77--116.
![Page 60: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/60.jpg)
SCALE-SPACE BLOB DETECTOR
1. Convolve image with scale-normalized Laplacian at several scales
![Page 61: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/61.jpg)
SCALE-SPACE BLOB DETECTOR: EXAMPLE
![Page 62: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/62.jpg)
SCALE-SPACE BLOB DETECTOR: EXAMPLE
![Page 63: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/63.jpg)
SCALE-SPACE BLOB DETECTOR
1. Convolve image with scale-normalized Laplacian at several scales
2. Find maxima of squared Laplacian response in scale-space
![Page 64: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/64.jpg)
SCALE-SPACE BLOB DETECTOR: EXAMPLE
![Page 65: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/65.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 66: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/66.jpg)
Approximating the Laplacian with a difference of Gaussians:
2 ( , , ) ( , , )xx yyL G x y G x y
( , , ) ( , , )DoG G x y k G x y
(Laplacian)
(Difference of Gaussians)
EFFICIENT IMPLEMENTATION
![Page 67: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/67.jpg)
EFFICIENT IMPLEMENTATION
David G. Lowe. "Distinctive image features from scale-invariant keypoints.” IJCV 60 (2), pp. 91-110, 2004.
![Page 68: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/68.jpg)
INVARIANCE AND COVARIANCE PROPERTIES
• Laplacian (blob) response is invariant w.r.t. rotation and scaling
• Blob location and scale is covariant w.r.t. rotation and scaling
• What about intensity change?
![Page 69: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/69.jpg)
OUTLINE
Corner detection Why detecting features? Finding corners: basic idea and mathematics Steps of Harris corner detector
Blob detection Scale selection Laplacian of Gaussian (LoG) detector Difference of Gaussian (DoG) detector Affine co-variant region
![Page 70: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/70.jpg)
ACHIEVING AFFINE COVARIANCE
• Affine transformation approximates viewpoint changes for roughly planar objects and roughly orthographic cameras
![Page 71: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/71.jpg)
ACHIEVING AFFINE COVARIANCE
RRIII
IIIyxwM
yyx
yxx
yx
2
112
2
, 0
0),(
direction of the slowest
change
direction of the fastest change
(max)-1/2
(min)-1/2
Consider the second moment matrix of the window containing the blob:
const][
v
uMvu
Recall:
This ellipse visualizes the “characteristic shape” of the window
![Page 72: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/72.jpg)
AFFINE ADAPTATION EXAMPLE
Scale-invariant regions (blobs)
![Page 73: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/73.jpg)
AFFINE ADAPTATION EXAMPLE
Affine-adapted blobs
![Page 74: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/74.jpg)
FROM COVARIANT DETECTION TO INVARIANT DESCRIPTION
• Geometrically transformed versions of the same neighborhood will give rise to regions that are related by the same transformation
• What to do if we want to compare the appearance of these image regions?
Normalization: transform these regions into same-size circles
![Page 75: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/75.jpg)
• Problem: There is no unique transformation from an ellipse to a unit circle
We can rotate or flip a unit circle, and it still stays a unit circle
AFFINE NORMALIZATION
![Page 76: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/76.jpg)
• To assign a unique orientation to circular image windows:
Create histogram of local gradient directions in the patch
Assign canonical orientation at peak of smoothed histogram
ELIMINATING ROTATION AMBIGUITY
0 2 p
![Page 77: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/77.jpg)
FROM COVARIANT REGIONS TO INVARIANT FEATURES
Extract affine regions Normalize regionsEliminate rotational
ambiguityCompute appearance
descriptors
SIFT (Lowe ’04)
![Page 78: CS 558 C OMPUTER V ISION Lecture VII: Corner and Blob Detection Slides adapted from S. Lazebnik.](https://reader035.fdocuments.us/reader035/viewer/2022062321/56649e8a5503460f94b8f929/html5/thumbnails/78.jpg)
INVARIANCE VS. COVARIANCE
Invariance: features(transform(image)) = features(image)
Covariance: features(transform(image)) =
transform(features(image))
Covariant detection => invariant description