Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 37

FACE RECOGNITION

USING 3D
TECHNIQUE
PRESENTED BY
INDU H MENON
EKAGEEC027
SEMINAR GUIDE:MISS.GINCY VARGHESE
OVERVIEW
 Introduction
 Face recognition
 3D face-expression recognition method
 New 3D face matching technique
 Surface matching
 PCA
 LDA
 3D face recognition algorithms
 Application
 Conclusion
 References
INTRODUCTION
 Face plays an important role in human communication. Facial
expressions and gestures incorporate nonverbal information which
contributes to human communication.
 With the development of 3D imaging technology, 3D face recognition
emerges as an alternative to overcome the difficulties inherent to 2D
face recognition, i.e. sensitivity to illumination conditions and positions
of a subject.
 It is composed of three subsystems: expression recognition system,
expressional face recognition system and neutral face recognition
system.
 A face recognition method that is able to recognize faces at various
angles is proposed. This method uses only the three dimensional range
images for matching.
 Firstly, surface matching, which consists of calculating the surface
distance of the probe face with the faces in the database, is performed
on the aligned face curves. The top ten candidates from the surface
matching are then further processed using Principal Component Analysis
(PCA) followed by Linear Discriminant Analysis (LDA).
BASIC STEPS IN FACE RECOGNITION

Input face image

Face detection

Face
Face feature
recognition
extraction

Face
database Feature Matching Decision maker

Output result
FACE DETECTION

 Geometric information based face detection


 Color information based face detection
 Combining them together

(b) Color
(a) Geometric
information based
information based
face detection
face detection
COLOR INFORMATION BASED
FACE DETECTION
 Face color is different from background
 Choice of color spaces is very important
Skin Background
color color

Color Spaces:
•R,G,B
•YCbCr
•YUV
•r,g
•……..
A FACE DETECTION ALGORITHM USING
COLOR AND GEOMETRIC INFORMATION
Ideas: (1) compensate for lightning, (2) separate by transforming to new
(sub) space.
Ideas: (1) compensate for lightning, (2) separate by
transforming to new (sub) space.
(3) clustering.
FEATURE-BASED FACE DETECTION

Color can be used in segmentation and grouping of image


subareas.
Location and shape parameters of eyes are the most important features
to be detected through segmentation and morphological operations
(dilation and erosion).
Ideas:
1) Eyes
2) Mouth
3) Boundary (edge detection)
4) Boundary approximated to
ellipse or something (Hough)
The concept of half-profiles
FACE MATCHING

 Feature based face matching


 Template matching
 Features versus templates
FEATURE BASED FACE MATCHING
You can extract
various features

Face image
From face Normalization Feature extraction
detection

Feature vector

Decision classifier
Output results
maker

You can use various You can use


decision makers various classifiers
NORMALIZATION

Eye
Normalization: rotation
location normalization, scale normalization

Averaged
for objects
Cross Correlation :
mean ( I T T )  mean ( I )( T )
C N (y ) 
 ( I T ) (T )

object template
FEATURE EXTRACTION
 Eyebrow thickness and vertical position at the eye center
position
 A coarse description of the left eyebrow’s arches
 Nose vertical position and width
 Mouth vertical position, width, height upper and lower lips
 eleven radii describing the chin shape
 Bigonial breadth (face width at nose position)
 Zygomatic breadth (face width halfway between nose tip
and eyes).

3.5-D feature vector


 Example of some
geometrical features
CLASSIFIER
This is just one example of
classifier, others are Decision Trees,
Bayes classifier expressions, decomposed structures,
NNs.


1
 j ( x)  ( x  mj ) T
( x  mj )

Rank the
distance
Feature vector Computer  (x) values
Output the
x j
results

 (x)
m
j

(j=2,3,…N)
j
ANN Classifier
Feature vector

Class 1 Class 2

ANN
one-class-in-one network
multi-class-in-one network
MAXNET

Classification results
TEMPLATE MATCHING
Templates
database You have
Produce a template to create
the data
base of
templates
Face image for all
From face Normalization matching
people
detection you want
to
recognize

Output results Decision


maker
WHY 3D FACE RECOGNITION?
3D FACE RECOGNITION
ALGORITHMS

You might also like