Face Detection: a Survey

advertisement

Face Detection: a Survey

Speaker: Mine-Quan Jing

National Chiao Tung University

Outline

Application

Related techniques

 Segmentation

 Identification

 Recognition

Progress ( 目前進展 )

Systems Demo

 NTU,NCTU,NTHU,ACADMIA

SINICA

The face detection techniques

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

The face detection techniques

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:

Segmentation

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 Based Segmentation

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:

Skin model construction

(Example)

The original image was taken from http://nn.csie.nctu.edu.tw/face-detection/ppframe.htm

Color based segmentation:

Skin model construction

(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

Change Detector

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 )

The face detection techniques

Image-Based Approach

 Linear Subspace Methods

 Neural Networks

 Statistical Approaches

Related News

The 5th International Conference on Automatic Face and Gesture

Recognition will take place 2002 in

Washington D.C., USA.

Download