1
CSE190a Fal 06
Face Recognition:
Fisherfaces and lighting
Biometrics
CSE 190a
Lecture 16
CSE190a Fal 06
Results of Hand Recognition!!
64%
Alex
42%
Warren (Nearest Neighbor)
32%
Warren (Bayesian)
60%
Taurin (Nearest Neighbor)
64%
Taurin (Statistical)
86%
Tom Nearest Neighbor
82%
Tom Bayesian
92%
Vikram (CS2) Nearest Neighbor
42%
Vikram (CS1)
Why is Face Recognition Hard?
CS252A, Winter 2005
Computer Vision I
Image as a Feature Vector
• Consider an npixel image to be a point in an
ndimensional space,
x
R
n
.
• Each pixel value is a coordinate of
x
.
∈
x
1
x
2
x
3
CS252A, Winter 2005
Computer Vision I
Nearest Neighbor Classifier
{
{
R
j
}
}
are set of training images.
x
1
x
2
x
3
R
1
R
2
I
)
,
(
min
arg
I
R
dist
ID
j
j
=
CS252A, Winter 2005
Computer Vision I
Comments
• Sometimes called “Template Matching”
• Variations on distance function (e.g. L
1
, robust
distances)
• Multiple templates per class perhaps many
training images per class.
• Expensive to compute k distances, especially
when each image is big (N dimensional).
• May not generalize well to unseen examples of
class.
• Some solutions:
– Bayesian classification
– Dimensionality reduction
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2
CS252A, Winter 2005
Computer Vision I
Appearancebased (Viewbased)
• Face Space:
– A set of face images construct a
face space
in
R
n
– Appearancebased methods analyze the distributions of
individual faces in face space
CS252A, Winter 2005
Computer Vision I
Eigenfaces: linear projection
•An
n
pixel image
x
∈
R
n
can be
projected to a lowdimensional
feature space
y
∈
R
m
by
y
= W
x
where
W
is an
n
by
m
matrix.
• Recognition is performed using
nearest neighbor in
R
m
.
• How do we choose a good
W
?
CS252A, Winter 2005
Computer Vision I
Eigenfaces: Principal Component Analysis (PCA)
Some details:
Use Singular value decomposition, “trick” described
in text to compute basis when
n<<d
CS252A, Winter 2005
Computer Vision I
How do you construct Eigenspace?
[
]
[
]
[
x
This is the end of the preview. Sign up
to
access the rest of the document.
 Fall '06
 Kriegman

Click to edit the document details