segmentation: mrfs and graph cuts computer vision cs 143, brown james hays 10/07/11 many slides from...
TRANSCRIPT
![Page 1: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/1.jpg)
![Page 2: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/2.jpg)
Segmentation: MRFs and Graph Cuts
Computer VisionCS 143, Brown
James Hays
10/07/11
Many slides from Kristin Grauman and Derek Hoiem
![Page 3: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/3.jpg)
Today’s class
• Segmentation and Grouping• Inspiration from human perception
– Gestalt properties• MRFs• Segmentation with Graph Cuts
i
j
wij
Slide: Derek Hoiem
![Page 4: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/4.jpg)
Grouping in vision• Goals:
– Gather features that belong together– Obtain an intermediate representation that compactly
describes key image or video parts
![Page 5: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/5.jpg)
Examples of grouping in vision
[Figure by J. Shi]
[http://poseidon.csd.auth.gr/LAB_RESEARCH/Latest/imgs/SpeakDepVidIndex_img2.jpg]
Determine image regions
Group video frames into shots
Fg / Bg
[Figure by Wang & Suter]
Object-level grouping
Figure-ground
[Figure by Grauman & Darrell]
Slide: Kristin Grauman
![Page 6: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/6.jpg)
Grouping in vision• Goals:
– Gather features that belong together– Obtain an intermediate representation that compactly
describes key image (video) parts
• Top down vs. bottom up segmentation– Top down: pixels belong together because they are from the
same object– Bottom up: pixels belong together because they look similar
• Hard to measure success– What is interesting depends on the app.
Slide: Kristin Grauman
![Page 7: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/7.jpg)
What things should be grouped?What cues indicate groups?
Slide: Kristin Grauman
![Page 8: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/8.jpg)
Gestalt psychology or Gestaltism
• German: Gestalt - "form" or "whole”• Berlin School, early 20th century
– Kurt Koffka, Max Wertheimer, and Wolfgang Köhler
• Gestalt: whole or group– Whole is greater than sum of its parts– Relationships among parts can yield
new properties/features• Psychologists identified series of factors
that predispose set of elements to be grouped (by human visual system)
![Page 9: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/9.jpg)
The Muller-Lyer illusion
Gestaltism
Slide: Derek Hoiem
![Page 10: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/10.jpg)
We perceive the interpretation, not the senses
Slide: Derek Hoiem
![Page 11: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/11.jpg)
Principles of perceptual organization
From Steve Lehar: The Constructive Aspect of Visual Perception
![Page 12: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/12.jpg)
Principles of perceptual organization
![Page 13: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/13.jpg)
Similarity
http://chicagoist.com/attachments/chicagoist_alicia/GEESE.jpg, http://wwwdelivery.superstock.com/WI/223/1532/PreviewComp/SuperStock_1532R-0831.jpg Slide: Kristin Grauman
![Page 14: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/14.jpg)
Symmetry
http://seedmagazine.com/news/2006/10/beauty_is_in_the_processingtim.php Slide: Kristin Grauman
![Page 15: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/15.jpg)
Common fate
Image credit: Arthus-Bertrand (via F. Durand)
Slide: Kristin Grauman
![Page 16: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/16.jpg)
Proximity
http://www.capital.edu/Resources/Images/outside6_035.jpg Slide: Kristin Grauman
![Page 17: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/17.jpg)
From Steve Lehar: The Constructive Aspect of Visual Perception
Grouping by invisible completion
![Page 18: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/18.jpg)
D. Forsyth
![Page 19: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/19.jpg)
![Page 20: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/20.jpg)
Emergence
![Page 21: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/21.jpg)
Gestalt cues
• Good intuition and basic principles for grouping
• Basis for many ideas in segmentation and occlusion reasoning
• Some (e.g., symmetry) are difficult to implement in practice
![Page 22: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/22.jpg)
intensity
pix
el c
ou
nt
input image
black pixelsgray pixels
white pixels
• These intensities define the three groups.• We could label every pixel in the image according to
which of these primary intensities it is.• i.e., segment the image based on the intensity feature.
• What if the image isn’t quite so simple?
1 23
Image segmentation: toy example
Kristen Grauman
![Page 23: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/23.jpg)
intensity
pix
el c
ou
nt
input image
input imageintensity
pix
el c
ou
nt
Kristen Grauman
![Page 24: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/24.jpg)
input imageintensity
pix
el c
ou
nt
• Now how to determine the three main intensities that define our groups?
• We need to cluster.
Kristen Grauman
![Page 25: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/25.jpg)
Clustering
• With this objective, it is a “chicken and egg” problem:– If we knew the cluster centers, we could allocate
points to groups by assigning each to its closest center.
– If we knew the group memberships, we could get the centers by computing the mean per group.
Kristen Grauman
![Page 26: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/26.jpg)
Smoothing out cluster assignments
• Assigning a cluster label per pixel may yield outliers:
1 23
?
original labeled by cluster center’s intensity
• How to ensure they are spatially smooth?
Kristen Grauman
![Page 27: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/27.jpg)
Solution
P(foreground | image)
Encode dependencies between pixels
edgesji
jiNi
i datayyfdatayfZ
dataP,
2..1
1 ),;,(),;(1
),;( y
Labels to be predicted Individual predictions Pairwise predictions
Normalizing constant
Slide: Derek Hoiem
![Page 28: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/28.jpg)
Writing Likelihood as an “Energy”
edgesji
jiNi
i datayypdataypZ
dataP,
2..1
1 ),;,(),;(1
),;( y
edgesji
jii
i datayydataydataEnergy,
21 ),;,(),;(),;( y
“Cost” of assignment yi
“Cost” of pairwise assignment yi ,yj
Slide: Derek Hoiem
![Page 29: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/29.jpg)
Markov Random Fields
edgesji
jii
i datayydataydataEnergy,
21 ),;,(),;(),;( y
Node yi: pixel label
Edge: constrained pairs
Cost to assign a label to each pixel
Cost to assign a pair of labels to connected pixels
Slide: Derek Hoiem
![Page 30: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/30.jpg)
Markov Random Fields
• Example: “label smoothing” gridUnary potential
0 10 0 K1 K 0
Pairwise Potential
0: -logP(yi = 0 ; data)1: -logP(yi = 1 ; data)
edgesji
jii
i datayydataydataEnergy,
21 ),;,(),;(),;( ySlide: Derek Hoiem
![Page 31: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/31.jpg)
Solving MRFs with graph cuts
edgesji
jii
i datayydataydataEnergy,
21 ),;,(),;(),;( y
Source (Label 0)
Sink (Label 1)
Cost to assign to 1
Cost to assign to 0
Cost to split nodes
Slide: Derek Hoiem
![Page 32: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/32.jpg)
Solving MRFs with graph cuts
edgesji
jii
i datayydataydataEnergy,
21 ),;,(),;(),;( y
Source (Label 0)
Sink (Label 1)
Cost to assign to 0
Cost to assign to 1
Cost to split nodes
Slide: Derek Hoiem
![Page 33: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/33.jpg)
GrabCut segmentation
User provides rough indication of foreground region.
Goal: Automatically provide a pixel-level segmentation.
Slide: Derek Hoiem
![Page 34: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/34.jpg)
Grab cuts and graph cutsGrab cuts and graph cuts Grab cuts and graph cutsGrab cuts and graph cuts
User Input
Result
Magic Wand (198?)
Intelligent ScissorsMortensen and Barrett (1995)
GrabCut
Regions Boundary Regions & Boundary
Source: Rother
![Page 35: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/35.jpg)
Colour ModelColour Model Colour ModelColour Model
Gaussian Mixture Model (typically 5-8 components)
Foreground &Background
Background G
R
Source: Rother
![Page 36: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/36.jpg)
Graph cutsGraph cuts Boykov and Jolly (2001)Boykov and Jolly (2001)
Graph cutsGraph cuts Boykov and Jolly (2001)Boykov and Jolly (2001)
ImageImage Min CutMin Cut
Cut: separating source and sink; Energy: collection of edges
Min Cut: Global minimal enegry in polynomial time
Foreground Foreground (source)(source)
BackgroundBackground(sink)(sink)
Source: Rother
![Page 37: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/37.jpg)
Colour ModelColour Model Colour ModelColour Model
Gaussian Mixture Model (typically 5-8 components)
Foreground &Background
Background
Foreground
BackgroundG
R
G
RIterated graph cut
Source: Rother
![Page 38: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/38.jpg)
GrabCut segmentation1. Define graph
– usually 4-connected or 8-connected2. Define unary potentials
– Color histogram or mixture of Gaussians for background and foreground
3. Define pairwise potentials
4. Apply graph cuts5. Return to 2, using current labels to compute
foreground, background models
2
2
21 2
)()(exp),(_
ycxc
kkyxpotentialedge
));((
));((log)(_
background
foreground
xcP
xcPxpotentialunary
Slide: Derek Hoiem
![Page 39: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/39.jpg)
What is easy or hard about these cases for graphcut-based segmentation?
Slide: Derek Hoiem
![Page 40: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/40.jpg)
Easier examples
GrabCut – Interactive Foreground Extraction 10
![Page 41: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/41.jpg)
More difficult Examples
Camouflage & Low Contrast
Harder CaseFine structure
Initial Rectangle
InitialResult
GrabCut – Interactive Foreground Extraction 11
![Page 42: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/42.jpg)
Lazy Snapping (Li et al. SG 2004)
![Page 43: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/43.jpg)
Using graph cuts for recognition
TextonBoost (Shotton et al. 2009 IJCV)
![Page 44: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/44.jpg)
Using graph cuts for recognition
TextonBoost (Shotton et al. 2009 IJCV)
Unary Potentials
Alpha Expansion Graph Cuts
![Page 45: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/45.jpg)
Limitations of graph cuts
• Associative: edge potentials penalize different labels
• If not associative, can sometimes clip potentials
• Approximate for multilabel– Alpha-expansion or alpha-beta swaps
Must satisfy
Slide: Derek Hoiem
![Page 46: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/46.jpg)
Graph cuts: Pros and Cons• Pros
– Very fast inference– Can incorporate data likelihoods and priors– Applies to a wide range of problems (stereo,
image labeling, recognition)• Cons
– Not always applicable (associative only)– Need unary terms (not used for generic
segmentation)• Use whenever applicable
Slide: Derek Hoiem
![Page 47: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/47.jpg)
More about MRFs/CRFs
• Other common uses– Graph structure on regions– Encoding relations between multiple scene
elements
• Inference methods– Loopy BP or BP-TRW: approximate, slower, but
works for more general graphs
Slide: Derek Hoiem
![Page 48: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/48.jpg)
Further reading and resources
• Graph cuts– http://www.cs.cornell.edu/~rdz/graphcuts.html– Classic paper: What Energy Functions can be Minimized via Graph Cuts?
(Kolmogorov and Zabih, ECCV '02/PAMI '04)
• Belief propagationYedidia, J.S.; Freeman, W.T.; Weiss, Y., "Understanding Belief Propagation and Its Generalizations”, Technical Report, 2001: http://www.merl.com/publications/TR2001-022/
• Normalized cuts and image segmentation (Shi and Malik)http://www.cs.berkeley.edu/~malik/papers/SM-ncut.pdf
• N-cut implementation http://www.seas.upenn.edu/~timothee/software/ncut/ncut.html
Slide: Derek Hoiem
![Page 49: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/49.jpg)
Next Class• Gestalt grouping
• More segmentation methods
![Page 50: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/50.jpg)
Recap of Grouping and Fitting
![Page 51: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/51.jpg)
Edge and line detection• Canny edge detector =
smooth derivative thin threshold link
• Generalized Hough transform = points vote for shape parameters
• Straight line detector = canny + gradient orientations orientation binning linking check for straightness
Slide: Derek Hoiem
![Page 52: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/52.jpg)
Robust fitting and registration
Key algorithms• RANSAC, Hough Transform
Slide: Derek Hoiem
![Page 53: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/53.jpg)
Clustering
Key algorithm• K-means
Slide: Derek Hoiem
![Page 54: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/54.jpg)
EM and Mixture of Gaussians
Tutorials: http://www.cs.duke.edu/courses/spring04/cps196.1/.../EM/tomasiEM.pdfhttp://www-clmc.usc.edu/~adsouza/notes/mix_gauss.pdf
Slide: Derek Hoiem
![Page 55: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/55.jpg)
Segmentation• Mean-shift segmentation
– Flexible clustering method, good segmentation
• Watershed segmentation– Hierarchical segmentation from soft boundaries
• Normalized cuts– Produces regular regions– Slow but good for oversegmentation
• MRFs with Graph Cut– Incorporates foreground/background/object
model and prefers to cut at image boundaries– Good for interactive segmentation or
recognition
Slide: Derek Hoiem
![Page 56: Segmentation: MRFs and Graph Cuts Computer Vision CS 143, Brown James Hays 10/07/11 Many slides from Kristin Grauman and Derek Hoiem](https://reader036.vdocument.in/reader036/viewer/2022062422/56649e935503460f94b98a3d/html5/thumbnails/56.jpg)
Next section: Recognition• How to recognize
– Specific object instances– Faces– Scenes– Object categories
Slide: Derek Hoiem