Beruflich Dokumente
Kultur Dokumente
Adjust to texture
2 2
MC2
MP2
PP3
MC3
MC4
PP4
MP3
MP4
MC5
PP5
MP5
PP2
Snakes
Data-driven
Prior knowledge
Deformable templates
Model learning
How can we learn what a bone looks like?
To build a statistical model of the data
we need many examples.
Objectives:
11
Example with
high probability
Example with
low probability
12
13
Height, Income
Height, Weight
Dimensionality
reduction by using
only leading
eigenvectors
18
Only first k
eigenvectors
++
21
22
Training set:
Instances vary in
- shape
- texture
- orientation
pose
- scale
- position
m Landmarks
represented as
1. Active Shape Models
23
s1
center of
gravity
scale
center
(translate)
rotate
24
to
by minimizing the
Procrustes distance
Recalculate the mean shape
from the aligned
set.
Iterate until the mean shape converges
25
26
ASM search
Initialize
Adjust to texture
1. Active Shape Models
28
ASM search
29 29
30
=
^
x
w1u1+w2u2+w3u3+w4u4+
31
Expression
Pose
Illumination
Long Term
Facial Hair
Makeup
Eyewear
Hairstyle
Piercings
Aging
Invariance Problem
Invariance Problem
D2 > 0
D3 ~
=0
D2 > 0
D3 > D1 or 2
[Hallinan 1994] [Adini, Moses, and Ullman 1994]
Bayesian classification
Dimensionality reduction
xn
x1
Lighting x Pose
x2
xn
x1
Real
Synthetic
[Georghiades, Belhumeur, and Kriegman 1999]
Eigenfaces example
Training images
x1,,xN
Eigenfaces example
Top eigenvectors: u1,uk
Mean:
Eigenfaces example
Principal component (eigenvector) uk
+ 3kuk
3kuk
Eigenfaces example
Face x in face space coordinates:
Eigenfaces example
Face x in face space coordinates:
=
Reconstruction:
=
^
x
w1u1+w2u2+w3u3+w4u4+
Expression
Pose
Illumination
Long Term
Facial Hair
Makeup
Eyewear
Hairstyle
Piercings
Aging
Limitations
Global appearance method: not robust to misalignment,
background variation
61
Expression
Pose
Illumination
Long Term
Facial Hair
Makeup
Eyewear
Hairstyle
Piercings
Aging
T emplate
S (X )
Ins tance
Shape-free texture
An attempt to eliminate texture variation
due to shape
Given instance I and a reference template
T we warp our image so that points of I
move into the corresponding points of T
I
warped to
becomes
AAM training
Training set: annotated grayscale images
Shape: displacements
Texture (shape-free)
PCA
Optional:
Interpolate to recover
dense deformation field
Model of shape
PCA
Model of texture
Appearance:
Synthesis:
I(S (x))
=
T (x)
X
T emplate
S (X )
Ins tance
AAM model
Shape Eigenvectors
Texture Eigenvectors
First four
modes of
appearance
variation
Find:
the best matching
synthetic image
How?
72
Variational Segmentation
E-step: segmentation
M-step: deformable model fitting
Atomic
Regions
M-step
E-step
Expression
Pose
Illumination
Long Term
Facial Hair
Makeup
Eyewear
Hairstyle
Piercings
Aging
3D Morphable models
Recover Shape
And then
82
83
84
88