MC949 - Computer Vision - INSTITUTO DE...
Transcript of MC949 - Computer Vision - INSTITUTO DE...
![Page 1: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/1.jpg)
MC949 - Computer VisionLecture #2
Microsoft Research Faculty FellowReasoning for Complex Data (RECOD) Lab.
Instituto de Computação, Universidade Estadual de Campinas (Unicamp)
[email protected]://www.ic.unicamp.br/~rocha
Prof. Dr. Anderson Rocha
![Page 2: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/2.jpg)
A. Rocha, 2012s1 – Computer Vision (MC949)
This lecture slides were made based on slides of several researchers such as James Hayes, Derek Hoiem, Alexei Efros, Steve Seitz, and David Forsyth and many others. Many thanks to all of these authors.
Projective Geometry and Camera Models
2
Reading: Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5
![Page 3: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/3.jpg)
A. Rocha, 2012s1 – Computer Vision (MC949)
Agenda
‣ Wrapping up last class
‣ Mapping between image and world coordinates
‣ Pinhole camera model (Building our own camera)
‣ Projective geometry
• Vanishing points and lines
‣ Projection matrix
3
![Page 4: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/4.jpg)
A. Rocha, 2012s1 – Computer Vision (MC949)
Last class: Intro
• Overview of vision, examples of state of art
• Computer Graphics: Models to Images
• Comp. Photography: Images to Images
• Computer Vision: Images to Models
![Page 5: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/5.jpg)
A. Rocha, 2012s1 – Computer Vision (MC949)
What do you need to make a camera from scratch?
![Page 6: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/6.jpg)
Today’s class: Camera and World
How tall is this woman?
Which ball is closer?
How high is the camera?
What is the camera rotation?
What is the focal length of the camera?
![Page 7: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/7.jpg)
Image formation
Let’s design a camera–Idea 1: put a piece of film in front of an object
–Do we get a reasonable image?
Slide source: Seitz
![Page 8: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/8.jpg)
Pinhole camera
Idea 2: add a barrier to block off most of the rays–This reduces blurring–The opening known as the aperture
Slide source: Seitz
![Page 9: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/9.jpg)
Pinhole camera
Figure from Forsyth
f
f = focal lengthc = center of the camera
c
![Page 10: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/10.jpg)
Camera obscura: the pre-camera• Known during classical period in China and
Greece (e.g. Mo-Ti, China, 470BC to 390BC)
Illustration of Camera Obscura Freestanding camera obscura at UNC Chapel Hill
Photo by Seth Ilys
![Page 11: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/11.jpg)
Camera Obscura used for Tracing
Lens Based Camera Obscura, 1568
![Page 12: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/12.jpg)
First Photograph
Oldest surviving photograph– Took 8 hours on pewter
plate
Joseph Niepce, 1826
Photograph of the first photograph
Stored at UT Austin
Niepce later teamed up with Daguerre, who eventually created Daguerrotypes
![Page 13: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/13.jpg)
Figures © Stephen E. Palmer, 2002
Dimensionality Reduction Machine (3D to 2D)
3D world 2D image
![Page 14: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/14.jpg)
Projection can be tricky…
Slide source: Seitz
![Page 15: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/15.jpg)
Projection can be tricky…
Slide source: Seitz
![Page 16: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/16.jpg)
Projective GeometryWhat is lost?• Length
Which is closer?
Who is taller?
![Page 17: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/17.jpg)
Length is not preserved
Figure by David Forsyth
B’
C
A’
![Page 18: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/18.jpg)
Projective GeometryWhat is lost?• Length• Angles
Perpendicular?
Parallel?
![Page 19: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/19.jpg)
Projective GeometryWhat is preserved?• Straight lines are still straight
![Page 20: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/20.jpg)
Vanishing points and lines Parallel lines in the world intersect in the
image at a “vanishing point”
![Page 21: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/21.jpg)
Vanishing points and lines
oVanishing Point oVanishing Point
Vanishing Line
![Page 22: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/22.jpg)
Vanishing points and lines
Vanishing point
Vanishing line
Vanishing point
Vertical vanishing point
(at infinity)
Slide from Efros, Photo from Criminisi
![Page 23: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/23.jpg)
Vanishing points and lines
Photo from online Tate collection
![Page 24: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/24.jpg)
Note on estimating vanishing points
![Page 25: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/25.jpg)
Projection: world coordinatesimage coordinates
Camera Center (tx,
ty, tz)
.
.. f Z Y.Optical Center (u0, v0)
v
u
![Page 26: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/26.jpg)
Homogeneous coordinates Conversion
Converting to homogeneous coordinates
homogeneous image coordinates
homogeneous scene coordinates
Converting from homogeneous coordinates
![Page 27: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/27.jpg)
Homogeneous coordinates
Invariant to scaling
Point in Cartesian is ray in Homogeneous
Homogeneous Coordinates
Cartesian Coordinates
![Page 28: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/28.jpg)
Basic geometry in homogeneous coordinates
• Line equation: ax + by + c = 0
• Append 1 to pixel coordinate to get homogeneous coordinate
• Line given by cross product of two points
• Intersection of two lines given by cross product of the lines
![Page 29: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/29.jpg)
Another problem solved by homogeneous coordinates
Cartesian: (Inf, Inf)Homogeneous: (1, 1, 0)
Intersection of parallel linesCartesian: (Inf, Inf) Homogeneous: (1, 2, 0)
![Page 30: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/30.jpg)
Slide Credit: SavereseProjection matrix
x: Image Coordinates: (u,v,1)K: Intrinsic Matrix (3x3)R: Rotation (3x3) t: Translation (3x1)X: World Coordinates: (X,Y,Z,1)
Ow
iw
kw
jwR,T
![Page 31: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/31.jpg)
Interlude: why does this matter?
![Page 32: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/32.jpg)
Object Recognition (CVPR 2006)
![Page 33: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/33.jpg)
Inserting photographed objects into images (SIGGRAPH 2007)
Original Created
![Page 34: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/34.jpg)
Pinhole Camera Model
x: Image Coordinates: (u,v,1)K: Intrinsic Matrix (3x3)R: Rotation (3x3) t: Translation (3x1)X: World Coordinates: (X,Y,Z,1)
![Page 35: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/35.jpg)
K
Slide Credit: Saverese
Projection matrix
Intrinsic Assumptions• Unit aspect ratio• Optical center at (0,0)• No skew
Extrinsic Assumptions• No rotation• Camera at (0,0,0)
![Page 36: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/36.jpg)
Remove assumption: known optical
Intrinsic Assumptions• Unit aspect ratio• No skew
Extrinsic Assumptions• No rotation• Camera at (0,0,0)
![Page 37: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/37.jpg)
Remove assumption: square pixels
Intrinsic Assumptions• No skew
Extrinsic Assumptions• No rotation• Camera at (0,0,0)
![Page 38: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/38.jpg)
Remove assumption: non-skewed pixels
Intrinsic Assumptions Extrinsic Assumptions• No rotation• Camera at (0,0,0)
Note: different books use different notation for parameters
![Page 39: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/39.jpg)
Oriented and Translated Camera
Ow
iw
kw
jw
t
R
![Page 40: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/40.jpg)
Allow camera translation
Intrinsic Assumptions Extrinsic Assumptions• No rotation
![Page 41: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/41.jpg)
3D Rotation of Points
Rotation around the coordinate axes, counter-clockwise:
p
p’γ
y
z
Slide Credit: Saverese
![Page 42: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/42.jpg)
Allow camera rotation
![Page 43: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/43.jpg)
Degrees of freedom
5 6
![Page 44: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/44.jpg)
Vanishing Point = Projection from Infinity
![Page 45: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/45.jpg)
Orthographic Projection• Special case of perspective projection
–Distance from the COP to the image plane is infinite
–Also called “parallel projection”–What’s the projection matrix?
Image World
Slide by Steve Seitz
![Page 46: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/46.jpg)
Scaled Orthographic Projection• Special case of perspective projection
–Object dimensions are small compared to distance to camera
–Also called “weak perspective”–What’s the projection matrix?
Image World
Slide by Steve Seitz
![Page 47: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/47.jpg)
Field of View (Zoom)
![Page 48: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/48.jpg)
Suppose we have two 3D cubes on the ground facing the viewer, one near, one far.
1.What would they look like in perspective?2.What would they look like in weak perspective?
Photo credit: GazetteLive.co.uk
![Page 49: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/49.jpg)
Beyond Pinholes: Radial Distortion
Image from Martin Habbecke
Corrected Barrel Distortion
![Page 50: MC949 - Computer Vision - INSTITUTO DE COMPUTAÇÃOrocha/teaching/2012s1/mc949/aulas/2012-vision... · MC949 - Computer Vision ... Szeliski, Chapter 2, Secs. 2.1 specially 2.1.5.](https://reader033.fdocuments.net/reader033/viewer/2022051801/5ae579527f8b9a6d4f8b5947/html5/thumbnails/50.jpg)
Things to remember
• Vanishing points and vanishing lines
• Pinhole camera model and camera projection matrix
• Homogeneous coordinates
Vanishing point
Vanishing line
Vanishing point
Vertical vanishing point
(at infinity)