single-view modeling, part 2
DESCRIPTION
CS4670/5670: Computer Vision. Noah Snavely. Single-view modeling, Part 2. Projective geometry. Readings Mundy, J.L. and Zisserman, A., Geometric Invariance in Computer Vision, Appendix: Projective Geometry for Machine Vision, MIT Press, Cambridge, MA, 1992, (read 23.1 - 23.5, 23.10) - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/1.jpg)
Single-view modeling, Part 2
CS4670/5670: Computer VisionNoah Snavely
![Page 2: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/2.jpg)
Projective geometry
• Readings– Mundy, J.L. and Zisserman, A., Geometric Invariance in Computer Vision, Appendix:
Projective Geometry for Machine Vision, MIT Press, Cambridge, MA, 1992, (read 23.1 - 23.5, 23.10)• available online: http://www.cs.cmu.edu/~ph/869/papers/zisser-mundy.pdf
Ames Room
![Page 3: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/3.jpg)
Announcements
• Midterm to be handed in Thursday by 5pm• Please hand it back at my office, Upson 4157
![Page 4: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/4.jpg)
Three point perspective
![Page 5: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/5.jpg)
Vanishing lines
• Multiple Vanishing Points– Any set of parallel lines on the plane define a vanishing
point– The union of all of these vanishing points is the horizon line
• also called vanishing line
– Note that different planes (can) define different vanishing lines
v1 v2
![Page 6: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/6.jpg)
Vanishing lines
• Multiple Vanishing Points– Any set of parallel lines on the plane define a vanishing
point– The union of all of these vanishing points is the horizon line
• also called vanishing line
– Note that different planes (can) define different vanishing lines
![Page 7: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/7.jpg)
Computing vanishing pointsV
DPP t 0
P0
D
![Page 8: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/8.jpg)
Computing vanishing points
• Properties– P is a point at infinity, v is its projection
– Depends only on line direction– Parallel lines P0 + tD, P1 + tD intersect at P
V
DPP t 0
0/1
/
/
/
1Z
Y
X
ZZ
YY
XX
ZZ
YY
XX
t D
D
D
t
t
DtP
DtP
DtP
tDP
tDP
tDP
PP
ΠPv
P0
D
![Page 9: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/9.jpg)
Computing vanishing lines
• Properties– l is intersection of horizontal plane through C with image plane– Compute l from two sets of parallel lines on ground plane– All points at same height as C project to l
• points higher than C project above l
– Provides way of comparing height of objects in the scene
ground plane
lC
v1 v2l
![Page 10: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/10.jpg)
![Page 11: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/11.jpg)
Fun with vanishing points
![Page 12: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/12.jpg)
Perspective cues
![Page 13: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/13.jpg)
Perspective cues
![Page 14: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/14.jpg)
Perspective cues
![Page 15: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/15.jpg)
Comparing heightsVanishingVanishing
PointPoint
![Page 16: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/16.jpg)
Measuring height
1
2
3
4
55.4
2.8
3.3
Camera heightHow high is the camera?
![Page 17: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/17.jpg)
q1
Computing vanishing points (from lines)
• Intersect p1q1 with p2q2
v
p1
p2
q2
Least squares version• Better to use more than two lines and compute the “closest” point of
intersection• See notes by Bob Collins for one good way of doing this:
– http://www-2.cs.cmu.edu/~ph/869/www/notes/vanishing.txt
![Page 18: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/18.jpg)
C
Measuring height without a ruler
ground plane
Compute Z from image measurements• Need more than vanishing points to do this
Z
![Page 19: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/19.jpg)
The cross ratio• A Projective Invariant– Something that does not change under projective
transformations (including perspective projection)
P1
P2
P3
P4
1423
2413
PPPP
PPPP
The cross-ratio of 4 collinear points
Can permute the point ordering• 4! = 24 different orders (but only 6 distinct values)
This is the fundamental invariant of projective geometry
1i
i
i
i Z
Y
X
P
3421
2431
PPPP
PPPP
![Page 20: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/20.jpg)
vZ
r
t
b
tvbr
rvbt
Z
Z
image cross ratio
Measuring height
B (bottom of object)
T (top of object)
R (reference point)
ground plane
HC
TBR
RBT
scene cross ratio
1
Z
Y
X
P
1
y
x
pscene points represented as image points as
R
H
R
H
R
![Page 21: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/21.jpg)
Measuring height
RH
vz
r
b
t
R
H
Z
Z
tvbr
rvbt
image cross ratio
H
b0
t0vvx vy
vanishing line (horizon)
![Page 22: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/22.jpg)
vz
r
b
t0vx vy
vanishing line (horizon)
v
t0
m0
What if the point on the ground plane b0 is not known?• Here the guy is standing on the box, height of box is known• Use one side of the box to help find b0 as shown above
b0
t1
b1
Measuring height
![Page 23: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/23.jpg)
3D Modeling from a photograph
St. Jerome in his Study, H. Steenwick
![Page 24: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/24.jpg)
3D Modeling from a photograph
![Page 25: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/25.jpg)
3D Modeling from a photograph
Flagellation, Piero della Francesca
![Page 26: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/26.jpg)
3D Modeling from a photograph
video by Antonio Criminisi
![Page 27: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/27.jpg)
3D Modeling from a photograph
![Page 28: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/28.jpg)
Camera calibration• Goal: estimate the camera parameters– Version 1: solve for projection matrix
ΠXx
1************
ZYX
wwywx
• Version 2: solve for camera parameters separately– intrinsics (focal length, principle point, pixel size)– extrinsics (rotation angles, translation)– radial distortion
![Page 29: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/29.jpg)
Vanishing points and projection matrix
************
Π 4321 ππππ
1π 2π 3π 4π
T00011 Ππ = vx (X vanishing point)
Z3Y2 , similarly, vπvπ
origin worldof projection10004 TΠπ
ovvvΠ ZYXNot So Fast! We only know v’s up to a scale factor
ovvvΠ ZYX cba• Can fully specify by providing 3 reference points
![Page 30: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/30.jpg)
Calibration using a reference object• Place a known object in the scene– identify correspondence between image and scene– compute mapping from scene to image
Issues• must know geometry very accurately• must know 3D->2D correspondence
![Page 31: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/31.jpg)
Chromaglyphs
Courtesy of Bruce Culbertson, HP Labshttp://www.hpl.hp.com/personal/Bruce_Culbertson/ibr98/chromagl.htm
![Page 32: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/32.jpg)
Estimating the projection matrix• Place a known object in the scene– identify correspondence between image and scene– compute mapping from scene to image
![Page 33: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/33.jpg)
Direct linear calibration
![Page 34: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/34.jpg)
Direct linear calibration
Can solve for mij by linear least squares• use eigenvector trick that we used for homographies
![Page 35: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/35.jpg)
Direct linear calibration• Advantage:– Very simple to formulate and solve
• Disadvantages:– Doesn’t tell you the camera parameters– Doesn’t model radial distortion– Hard to impose constraints (e.g., known f)– Doesn’t minimize the right error function
For these reasons, nonlinear methods are preferred• Define error function E between projected 3D points and image positions
– E is nonlinear function of intrinsics, extrinsics, radial distortion
• Minimize E using nonlinear optimization techniques
![Page 36: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/36.jpg)
Alternative: multi-plane calibration
Images courtesy Jean-Yves Bouguet, Intel Corp.
Advantage• Only requires a plane• Don’t have to know positions/orientations• Good code available online! (including in OpenCV)
– Matlab version by Jean-Yves Bouget: http://www.vision.caltech.edu/bouguetj/calib_doc/index.html
– Zhengyou Zhang’s web site: http://research.microsoft.com/~zhang/Calib/
![Page 37: Single-view modeling, Part 2](https://reader035.vdocuments.net/reader035/viewer/2022062321/568139c9550346895da1773a/html5/thumbnails/37.jpg)
Some Related Techniques• Image-Based Modeling and Photo Editing– Mok et al., SIGGRAPH 2001– http://graphics.csail.mit.edu/ibedit/
• Single View Modeling of Free-Form Scenes– Zhang et al., CVPR 2001– http://grail.cs.washington.edu/projects/svm/
• Tour Into The Picture– Anjyo et al., SIGGRAPH 1997– http://koigakubo.hitachi.co.jp/little/DL_TipE.html