Speaker: Mine-Quan Jing
National Chiao Tung University
Application
Related techniques
Segmentation
Identification
Recognition
Progress ( 目前進展 )
Systems Demo
NTU,NCTU,NTHU,ACADMIA
SINICA
Feature-Based Approach
Skin color and face geometry
Detection task is accomplished by
Distance, angles and area of visual features
Image-Based Approach
As a general recognition system
Feature-Based Approach
Low-Level Analysis
Segmentation of visual features
Feature Analysis
Organized the features into
1. Global concept
2. Facial features
Active Shape Models
Extract the complex & non-rigid feature
Ex: eye pupil, lip tracking.
Low-Level Analysis:
of visual features
Edges: (The most primitive feature)
Trace a human head outline.
Provide the information
Shape & position of the face
Edge operators
Sobel
Marr-Hildreth
first and second derivatives of
Gaussians
Low-Level Analysis:
Segmentation of visual features
The steerable filtering
1. Detection of edges
2. Determining the orientation
3. Tracking the neighboring edges
Edge-detection system
1.
2.
3.
1. Label the edge
2. Matched to a face model
3. Golden ratio height
1
width 2
5
Low-Level Analysis:
Segmentation of visual features
Gray information
Facial feature ( eyebrows , pupils … )
Darker than their surrounding
Application
Search an eye pair
Find the bright pixel (nose tips)
Mosaic (pyramid) images
Segmentation of visual features:
Color information
Difference races?
Different skin color gives rise to a tight cluster in color space .
Color models
Normalized RGB colors
A color histogram for a face is made r g
R
R
R
G
G
G
B
B b
R
B
G
B
Comparing the color of a pixel with respect to the r and g.
Why normalized ? Brightness change
Low-Level Analysis:
Segmentation of visual features
HSI color model
For large variance among facial feature clusters [106].
• Extract lips, eyes, and eyebrows .
Also used in face segmentation
YIQ
Color ’ s ranging from orange to cyan
• Enhance the skin region of Asians [29].
Other color models
HSV, YES, CIE-xyz …
Comparative study of color space [Terrilon
188]
Low-Level Analysis:
Segmentation of visual features
Color segmentation by color thresholds
Skin color is modeled through
Histogram or charts (simple)
Statistical measures (complex)
Ex:
• Skin color cluster can be represented as Gaussian distribution [215]
Advantage of Statistical color model
The model is updatable
More robust against changes in environment
Low-Level Analysis:
Segmentation of visual features
The disadvantage:
Not robust under varying lighting condiction
Color based segmentation:
(Example)
The original image was taken from http://nn.csie.nctu.edu.tw/face-detection/ppframe.htm
Color based segmentation:
(Example)
The original image was taken from http://nn.csie.nctu.edu.tw/face-detection/ppframe.htm
Low-Level Analysis:
Segmentation of visual features
Motion information a face is almost always moving
Disadvantages:
What if there are other object moving in the background.
1.
Four steps for detection
Frame differencing
2.
3.
4.
Thresholding
Noise removal
Locate the face http://ansatte.hig.no/~erikh/papers/hig98_6/node2.html#bevdet
Related techniques
Amount of pixels on each line in the motion image
A typical motion image
The original images were taken from http://ansatte.hig.no/~erikh/papers/hig98_6/node2.html#bevdet
Motion-Based segmentation:
Motion estimation [126]
People are always moving.
For focusing of attention
discard cluttered, static background
A spatio-temporal Gaussian filter can be used to detect moving boundaries of faces.
G ( x , y , t ) m ( x , y , t )
a u (
) 2
3 e
a ( x
2 y
2 u
2 t
2
)
2
1 u
2
2
t
2
G ( x , y , t )
Image-Based Approach
Linear Subspace Methods
Neural Networks
Statistical Approaches
The 5th International Conference on Automatic Face and Gesture
Recognition will take place 2002 in
Washington D.C., USA.