fitting a transformation feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · fitting...
TRANSCRIPT
![Page 1: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/1.jpg)
4/28/20
1
Fittingatransformation:Feature-basedalignment
April28th,2020
YongJaeLeeUCDavis
GeneralizedHoughTransform
Modelimage VotespaceNovelimage
xxx
xx
Nowsupposethosecolorsencodegradientdirections…
• Whatifwewanttodetectarbitraryshapes?
Intuition:
Ref.point
Displacementvectors
KristenGrauman2
• Defineamodelshapebyitsboundarypointsandareferencepoint.
[Dana H. Ballard, Generalizing the Hough Transform to Detect Arbitrary Shapes, 1980]
x
a
p1
θ p2 θ
At each boundary point, compute displacement vector: r = a – pi. Store these vectors in a table indexed by gradient orientation θ.
GeneralizedHoughTransform
Offlineprocedure:
Modelshape
θ
θ
…
…
…
3
![Page 2: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/2.jpg)
4/28/20
2
p1
θ θ
Foreachedgepoint:• Useitsgradientorientationθto
indexintostoredtable
• Useretrievedrvectorstovoteforreferencepoint
GeneralizedHoughTransform
Detectionprocedure:
Assuming translation is the only transformation here, i.e., orientation and scale are fixed.
x
θ θ
Novelimage
θ
θ
…
…
…
θ
xx
xx
4
GeneralizedHoughforobjectdetection
• Insteadofindexingdisplacementsbygradientorientation,indexbymatchedlocalpatterns.
B. Leibe, A. Leonardis, and B. Schiele, Combined Object Categorization and Segmentation with an Implicit Shape Model, ECCV Workshop on Statistical Learning in Computer Vision 2004
training image
“visual codeword” with displacement vectors
Source: L. Lazebnik 5
• Insteadofindexingdisplacementsbygradientorientation,indexby“visualcodeword”
B. Leibe, A. Leonardis, and B. Schiele, Combined Object Categorization and Segmentation with an Implicit Shape Model, ECCV Workshop on Statistical Learning in Computer Vision 2004
test image
Source: L. Lazebnik
GeneralizedHoughforobjectdetection
6
![Page 3: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/3.jpg)
4/28/20
3
Summary• Fittingproblemsrequirefindinganysupportingevidencefora
model,evenwithinclutterandmissingfeatures– associatefeatureswithanexplicitmodel
• Votingapproaches,suchastheHoughtransform,makeitpossibletofindlikelymodelparameterswithoutsearchingallcombinationsoffeatures– Houghtransformapproachforlines,circles,…,arbitraryshapes
definedbyasetofboundarypoints,recognitionfrompatches
7
Today
• Feature-based alignment – 2D transformations – Affine fit – RANSAC
8
Motivation: Recognition
Figures from David Lowe 9
![Page 4: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/4.jpg)
4/28/20
4
Motivation: medical image registration
10
Slide credit: Kristen Grauman
Motivation: mosaics
Image from http://graphics.cs.cmu.edu/courses/15-463/2010_fall/
(In detail next week)
11
Alignment problem • We have previously considered how to fit a model to
image evidence – e.g., a line to edge points
• In alignment, we will fit the parameters of some transformation according to a set of matching feature pairs (“correspondences”).
T
xi
xi '
12
Slide credit: Kristen Grauman
![Page 5: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/5.jpg)
4/28/20
5
Parametric (global) warping Examples of parametric warps:
translation rotation aspect
affine perspective
Source: Alyosha Efros 13
Parametric (global) warping
Transformation T is a coordinate-changing machine: p’ = T(p)
What does it mean that T is global? • Is the same for any point p • can be described by just a few numbers (parameters)
Let’s represent T as a matrix: p’ = Mp
T
p = (x,y) p’ = (x’,y’)
⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡
yx
yx
M''
Source: Alyosha Efros 14
Scaling Scaling a coordinate means multiplying each of its components by
a scalar Uniform scaling means this scalar is the same for all components:
× 2
Source: Alyosha Efros
15
![Page 6: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/6.jpg)
4/28/20
6
Non-uniform scaling: different scalars per component:
Scaling
X × 2, Y × 0.5
Source: Alyosha Efros
16
Scaling
Scaling operation: Or, in matrix form:
byyaxx
=
=
''
⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡
yx
ba
yx
00
''
scaling matrix S
Source: Alyosha Efros
17
What transformations can be represented with a 2x2 matrix?
2D Rotate around (0,0)?
yxyyxx*cos*sin'*sin*cos'
Θ+Θ=Θ−Θ=
⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡
ΘΘ
Θ−Θ=⎥
⎦
⎤⎢⎣
⎡
yx
yx
cossinsincos
''
2D Shear?
yxshyyshxx
y
x
+=
+=
*'*'
⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡
yx
shsh
yx
y
x
11
''
Source: Alyosha Efros
2D Scaling?
ysyxsx
y
x
*'
*'
=
=⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡
yx
ss
yx
y
x
00
''
18
![Page 7: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/7.jpg)
4/28/20
7
What transformations can be represented with a 2x2 matrix?
Source: Alyosha Efros
2D Mirror about Y axis?
yyxx
=−=
''
⎥⎦⎤
⎢⎣⎡⎥⎦⎤
⎢⎣⎡−=⎥⎦
⎤⎢⎣⎡
yx
yx
1001
''
2D Mirror over (0,0)?
yyxx−=−=
''
⎥⎦⎤
⎢⎣⎡⎥⎦⎤
⎢⎣⎡
−−=⎥⎦
⎤⎢⎣⎡
yx
yx
1001
''
2D Translation?
y
x
tyytxx
+=
+=
''
NO! 19
2D Linear Transformations
Only linear 2D transformations can be represented with a 2x2 matrix.
Linear transformations are combinations of … • Scale, • Rotation, • Shear, and • Mirror
⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡
yx
dcba
yx''
Source: Alyosha Efros 20
Homogeneous coordinates
Converting from homogeneous coordinates:
homogeneous image coordinates
To convert to homogeneous coordinates:
21
Slide credit: Kristen Grauman
Convenient coordinate system to represent many useful transformations
![Page 8: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/8.jpg)
4/28/20
8
Homogeneous Coordinates Q: How can we represent 2d translation as a 3x3 matrix
using homogeneous coordinates? A: Using the rightmost column:
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
=
1001001
y
x
tt
ranslationT
y
x
tyytxx
+=
+=
''
Source: Alyosha Efros 22
Translation
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
+
+
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
111001001
1''
y
x
y
x
tytx
yx
tt
yx
tx = 2 ty = 1
Homogeneous Coordinates
Source: Alyosha Efros
23
Basic 2D Transformations Basic 2D transformations as 3x3 matrices
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
ΘΘ
Θ−Θ
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
11000cossin0sincos
1''
yx
yx
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
11001001
1''
yx
tt
yx
y
x
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
11000101
1''
yx
shsh
yx
y
x
Translate
Rotate Shear
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
11000000
1''
yx
ss
yx
y
x
Scale
Source: Alyosha Efros 24
![Page 9: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/9.jpg)
4/28/20
9
2D Affine Transformations
Affine transformations are combinations of … • Linear transformations, and • Translations
Parallel lines remain parallel
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
=
⎥⎥⎥
⎦
⎤
⎢⎢⎢
⎣
⎡
wyx
fedcba
wyx
100'''
25
Slide credit: Kristen Grauman
Today
• Feature-based alignment – 2D transformations – Affine fit – RANSAC
26
Alignment problem • We have previously considered how to fit a model to
image evidence – e.g., a line to edge points
• In alignment, we will fit the parameters of some transformation according to a set of matching feature pairs (“correspondences”).
T
xi
xi '
Kristen Grauman
27
![Page 10: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/10.jpg)
4/28/20
10
Image alignment
• Two broad approaches: – Direct (pixel-based) alignment
• Search for alignment where most pixels agree – Feature-based alignment
• Search for alignment where extracted features agree • Can be verified using pixel-based alignment 28
Slide credit: Kristen Grauman
Fitting an affine transformation • Assuming we know the correspondences, how do we
get the transformation?
),( ii yx ʹʹ),( ii yx
⎥⎦
⎤⎢⎣
⎡+⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡ʹ
ʹ
2
1
43
21
tt
yx
mmmm
yx
i
i
i
i
29
Slide credit: Kristen Grauman
An aside: Least Squares Example Say we have a set of data points (X1,X1’), (X2,X2’),
(X3,X3’), etc. (e.g. person’s height vs. weight) We want a nice compact formula (a line) to predict X’s
from Xs: Xa + b = X’ We want to find a and b How many (X,X’) pairs do we need?
What if the data is noisy?
'22
'11
XbaXXbaX
=+
=+⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡'2
'1
2
1
11
XX
ba
XX
Ax=B
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
=⎥⎦
⎤⎢⎣
⎡
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
.........111
'3
'2
'1
3
2
1
XXX
ba
XXX
overconstrained
2min BAx −
Source: Alyosha Efros
30
![Page 11: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/11.jpg)
4/28/20
11
Fitting an affine transformation • Assuming we know the correspondences, how do we
get the transformation?
),( ii yx ʹʹ),( ii yx
⎥⎦
⎤⎢⎣
⎡+⎥⎦
⎤⎢⎣
⎡⎥⎦
⎤⎢⎣
⎡=⎥
⎦
⎤⎢⎣
⎡ʹ
ʹ
2
1
43
21
tt
yx
mmmm
yx
i
i
i
i
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
ʹ
ʹ=
⎥⎥⎥⎥⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢⎢⎢⎢⎢
⎣
⎡
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
!
!
!
!
i
i
ii
ii
yx
ttmmmm
yxyx
2
1
4
3
2
1
10000100
31
Slide credit: Kristen Grauman
Fitting an affine transformation
• How many matches (correspondence pairs) do we need to solve for the transformation parameters?
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
ʹ
ʹ=
⎥⎥⎥⎥⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢⎢⎢⎢⎢
⎣
⎡
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
!
!
!
!
i
i
ii
ii
yx
ttmmmm
yxyx
2
1
4
3
2
1
10000100
Kristen Grauman
32
Affine: # correspondences?
How many correspondences needed for affine?
x x’
T(x,y) y y’
?
Alyosha Efros
![Page 12: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/12.jpg)
4/28/20
12
Fitting an affine transformation
• How many matches (correspondence pairs) do we need to solve for the transformation parameters?
• Once we have solved for the parameters, how do we compute the coordinates of the corresponding point for ?
• Where do the matches come from?
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
ʹ
ʹ=
⎥⎥⎥⎥⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢⎢⎢⎢⎢
⎣
⎡
⎥⎥⎥⎥
⎦
⎤
⎢⎢⎢⎢
⎣
⎡
!
!
!
!
i
i
ii
ii
yx
ttmmmm
yxyx
2
1
4
3
2
1
10000100
),( newnew yx
Kristen Grauman
34
What are the correspondences?
?
• Compare content in local patches, find best matches. e.g., simplest approach: scan with template, and compute SSD or correlation between list of pixel intensities in the patch
• Later in the course: how to select regions using more robust descriptors.
Kristen Grauman
35
Fitting an affine transformation
Figures from David Lowe, ICCV 1999 36
![Page 13: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/13.jpg)
4/28/20
13
37
Fitting an affine transformation
38
Fitting an affine transformation
Fitting an affine transformation
Figures from David Lowe, ICCV 1999 39
![Page 14: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/14.jpg)
4/28/20
14
Today
• Feature-based alignment – 2D transformations – Affine fit – RANSAC
40
Outliers • Outliers can hurt the quality of our parameter
estimates, e.g., – an erroneous pair of matching points from two images – an edge point that is noise, or doesn’t belong to the
line we are fitting.
Kristen Grauman
41
Outliers affect least squares fit
42
Slide credit: Kristen Grauman
![Page 15: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/15.jpg)
4/28/20
15
Outliers affect least squares fit
43
Slide credit: Kristen Grauman
RANSAC
• RANdom Sample Consensus
• Approach: we want to avoid the impact of outliers, so let’s look for “inliers”, and use those only.
• Intuition: if an outlier is chosen to compute the current fit, then the resulting line won’t have much support from rest of the points.
44
Slide credit: Kristen Grauman
RANSAC: General form • RANSAC loop: 1. Randomly select a seed group of points on which to
base transformation estimate
2. Compute transformation from seed group
3. Find inliers to this transformation
4. If the number of inliers is sufficiently large, re-compute estimate of transformation on all of the inliers
• Keep the transformation with the largest number of inliers
45
Slide credit: Kristen Grauman
![Page 16: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/16.jpg)
4/28/20
16
RANSAC for line fitting example
Source: R. Raguram Lana Lazebnik
46
RANSAC for line fitting example
Least-squaresfit
Source: R. Raguram Lana Lazebnik
47
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
Source: R. Raguram Lana Lazebnik
48
![Page 17: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/17.jpg)
4/28/20
17
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
Source: R. Raguram Lana Lazebnik
49
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
3. Computeerrorfunction
Source: R. Raguram Lana Lazebnik
50
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
3. Computeerrorfunction
4. Selectpointsconsistentwithmodel
Source: R. Raguram Lana Lazebnik
51
![Page 18: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/18.jpg)
4/28/20
18
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
3. Computeerrorfunction
4. Selectpointsconsistentwithmodel
5. Repeathypothesize-and-verifyloop
Source: R. Raguram Lana Lazebnik
52
53
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
3. Computeerrorfunction
4. Selectpointsconsistentwithmodel
5. Repeathypothesize-and-verifyloop
Source: R. Raguram Lana Lazebnik
54
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
3. Computeerrorfunction
4. Selectpointsconsistentwithmodel
5. Repeathypothesize-and-verifyloop
Source: R. Raguram Lana Lazebnik
![Page 19: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/19.jpg)
4/28/20
19
RANSAC for line fitting example
1. Randomlyselectminimalsubsetofpoints
2. Hypothesizeamodel
3. Computeerrorfunction
4. Selectpointsconsistentwithmodel
5. Repeathypothesize-and-verifyloop
Source: R. Raguram Lana Lazebnik
55
RANSAC for line fitting Repeat N times: • Draw s points uniformly at random • Fit line to these s points • Find inliers to this line among the remaining
points (i.e., points whose distance from the line is less than t)
• If there are d or more inliers, accept the line and refit using all inliers
Lana Lazebnik
56
RANSAC pros and cons • Pros
• Simple and general • Applicable to many different problems • Often works well in practice
• Cons • Lots of parameters to tune • Doesn’t work well for low inlier ratios (too many iterations,
or can fail completely)
Lana Lazebnik
57
![Page 20: Fitting a transformation Feature-based alignmentyjlee/teaching/ecs174-spring2020/lee... · Fitting a transformation: Feature-based alignment April 28th, 2020 Yong Jae Lee UC Davis](https://reader033.vdocuments.us/reader033/viewer/2022043006/5f8ef7a422e07a74a4068f0b/html5/thumbnails/20.jpg)
4/28/20
20
Today
• Feature-based alignment – 2D transformations – Affine fit – RANSAC
58
Comingup:alignmentandimagestitching
59
Questions?
SeeyouThursday!
60