OBJ%20CUT

1
OBJ CUT
UNIVERSITY OF OXFORD

• M. Pawan Kumar
• Philip Torr
• Andrew Zisserman

2
Aim

• Given an image, to segment the object

Object Category Model
Segmentation
Cow Image
Segmented Cow

• Segmentation should (ideally) be
• shaped like the object e.g. cow-like
• obtained efficiently in an unsupervised manner
• able to handle self-occlusion

3
Challenges
Intra-Class Shape Variability
Intra-Class Appearance Variability
Self Occlusion
4
Motivation
Magic Wand

• Current methods require user intervention
• Object and background seed pixels (Boykov and
  Jolly, ICCV 01)
• Bounding Box of object (Rother et al. SIGGRAPH
  04)

Object Seed Pixels
Cow Image
5
Motivation
Magic Wand

• Current methods require user intervention
• Object and background seed pixels (Boykov and
  Jolly, ICCV 01)
• Bounding Box of object (Rother et al. SIGGRAPH
  04)

Object Seed Pixels
Background Seed Pixels
Cow Image
6
Motivation
Magic Wand

• Current methods require user intervention
• Object and background seed pixels (Boykov and
  Jolly, ICCV 01)
• Bounding Box of object (Rother et al. SIGGRAPH
  04)

Segmented Image
7
Motivation
Magic Wand

• Current methods require user intervention
• Object and background seed pixels (Boykov and
  Jolly, ICCV 01)
• Bounding Box of object (Rother et al. SIGGRAPH
  04)

Object Seed Pixels
Background Seed Pixels
Cow Image
8
Motivation
Magic Wand

• Current methods require user intervention
• Object and background seed pixels (Boykov and
  Jolly, ICCV 01)
• Bounding Box of object (Rother et al. SIGGRAPH
  04)

Segmented Image
9
Motivation

• Problem
• Manually intensive
• Segmentation is not guaranteed to be
  object-like

Non Object-like Segmentation
10
Our Method

• Combine object detection with segmentation
• Borenstein and Ullman, ECCV 02
• Leibe and Schiele, BMVC 03
• Incorporate global shape priors in MRF
• Detection provides
• Object Localization
• Global shape priors
• Automatically segments the object
• Note our method completely generic
• Applicable to any object category model

11
Outline

• Problem Formulation
• Form of Shape Prior
• Optimization
• Results

12
Problem

• Labelling m over the set of pixels D
• Shape prior provided by parameter T
• Energy E (m,T) ?Fx(Dmx)Fx(mxT) ?
  ?xy(mx,my) F(Dmx,my)
• Unary terms
• Likelihood based on colour
• Unary potential based on distance from T
• Pairwise terms
• Prior
• Contrast term
• Find best labelling m arg min ? wi E (m,Ti)
• wi is the weight for sample Ti

Unary terms
Pairwise terms
13
MRF

• Probability for a labelling consists of
• Likelihood
• Unary potential based on colour of pixel
• Prior which favours same labels for neighbours
  (pairwise potentials)

mx
m (labels)
Prior ?xy(mx,my)
my
Unary Potential Fx(Dmx)
x
y
D (pixels)
Image Plane
14
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Fx(Dobj)
x

x

Fx(Dbkg)
?xy(mx,my)
y

y





Prior
Likelihood Ratio (Colour)
15
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Prior
Likelihood Ratio (Colour)
16
Contrast-Dependent MRF

• Probability of labelling in addition has
• Contrast term which favours boundaries to lie on
  image edges

mx
m (labels)
my
x
Contrast Term F(Dmx,my)
y
D (pixels)
Image Plane
17
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Fx(Dobj)
x

x

Fx(Dbkg)
?xy(mx,my) F(Dmx,my)
y

y





Prior Contrast
Likelihood Ratio (Colour)
18
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Prior Contrast
Likelihood Ratio (Colour)
19
Our Model

• Probability of labelling in addition has
• Unary potential which depend on distance from T
  (shape parameter)

T (shape parameter)
Unary Potential Fx(mxT)
mx
m (labels)
my
Object Category Specific MRF
x
y
D (pixels)
Image Plane
20
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Shape Prior T
Prior Contrast
Distance from T
21
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Shape Prior T
Prior Contrast
Likelihood Distance from T
22
Example
Cow Image
Object Seed Pixels
Background Seed Pixels
Shape Prior T
Prior Contrast
Likelihood Distance from T
23
Outline

• Problem Formulation
• E (m,T) ?Fx(Dmx)Fx(mxT) ? ?xy(mx,my)
  F(Dmx,my)
• Form of Shape Prior
• Optimization
• Results

24
Layered Pictorial Structures (LPS)

• Generative model
• Composition of parts spatial layout

Layer 2
Spatial Layout (Pairwise Configuration)
Layer 1
Parts in Layer 2 can occlude parts in Layer 1
25
Layered Pictorial Structures (LPS)
Cow Instance
Layer 2
Transformations
T1 P(T1) 0.9
Layer 1
26
Layered Pictorial Structures (LPS)
Cow Instance
Layer 2
Transformations
T2 P(T2) 0.8
Layer 1
27
Layered Pictorial Structures (LPS)
Unlikely Instance
Layer 2
Transformations
T3 P(T3) 0.01
Layer 1
28
LPS for Detection

• Learning
• Learnt automatically using a set of examples
• Detection
• Matches LPS to image using Loopy Belief
  Propagation
• Localizes object parts

29
Outline

• Problem Formulation
• Form of Shape Prior
• Optimization
• Results

30
Optimization

• Given image D, find best labelling as
  m arg max p(mD)
• Treat LPS parameter T as a latent (hidden)
  variable
• EM framework
• E sample the distribution over T
• M obtain the labelling m

31
E-Step

• Given initial labelling m, determine p(Tm,D)
• Problem
• Efficiently sampling from p(Tm,D)
• Solution
• We develop efficient sum-product Loopy Belief
  Propagation (LBP) for matching LPS.
• Similar to efficient max-product LBP for MAP
  estimate
• Felzenszwalb and Huttenlocher, CVPR 04

32
Results

• Different samples localize different parts well.
• We cannot use only the MAP estimate of the LPS.

33
M-Step

• Given samples from p(Tm,D), get new labelling
  mnew
• Sample Ti provides
• Object localization to learn RGB distributions of
  object and background
• Shape prior for segmentation
• Problem
• Maximize expected log likelihood using all
  samples
• To efficiently obtain the new labelling

34
M-Step
w1 P(T1m,D)
Cow Image
Shape T1
RGB Histogram for Background
RGB Histogram for Object
35
M-Step
w1 P(T1m,D)
Cow Image
Shape T1
T1
m (labels)
Image Plane
D (pixels)

• Best labelling found efficiently using a Single
  Graph Cut

36
Segmentation using Graph Cuts
Obj
Cut
Fx(Dbkg) Fx(bkgT)
x

?xy(mx,my) F(Dmx,my)
y



m
z


Fz(Dobj) Fz(objT)
Bkg
37
Segmentation using Graph Cuts
Obj
x

y



m
z


Bkg
38
M-Step
w2 P(T2m,D)
Cow Image
Shape T2
RGB Histogram for Background
RGB Histogram for Object
39
M-Step
w2 P(T2m,D)
Cow Image
Shape T2
T2
m (labels)
Image Plane
D (pixels)

• Best labelling found efficiently using a Single
  Graph Cut

40
M-Step
T2
T1
w1
w2
.
Image Plane
Image Plane
m arg min ? wi E (m,Ti)

• Best labelling found efficiently using a Single
  Graph Cut

41
Outline

• Problem Formulation
• Form of Shape Prior
• Optimization
• Results

42
Results
Using LPS Model for Cow
Segmentation
Image
43
Results
Using LPS Model for Cow
In the absence of a clear boundary between object
and background
Segmentation
Image
44
Results
Using LPS Model for Cow
Segmentation
Image
45
Results
Using LPS Model for Cow
Segmentation
Image
46
Results
Using LPS Model for Horse
Segmentation
Image
47
Results
Using LPS Model for Horse
Segmentation
Image
48
Results
Our Method
Leibe and Schiele
Image
49
Results
Shape
ShapeAppearance
Appearance
Without Fx(mxT)
Without Fx(Dmx)
50

• Conclusions
• New model for introducing global shape prior in
  MRF
• Method of combining detection and segmentation
• Efficient LBP for detecting articulated objects
• Future Work
• Other shape parameters need to be explored
• Method needs to be extended to handle multiple
  visual aspects