Principal Component Analysis Using Eigenfaces for Face Recognition

8/2/2019 Principal Component Analysis Using Eigenfaces for Face Recognition

1/18

Principal Component Analysis usingEigenfaces for face recognition

&Viola-Jones method for face

detection


2/18


3/18

WHY SELECT 'PCA based Eigenfaces' method?

PCA based Eigenface method is at the most primary

level and simplest ofefficient face recognition

algorithms --- and is therefore a great place for

beginners to start learning face recognition!

PCA based Eigenfaces method for recognition is as

supported by EmguCV library as is Viola-Jones

method for detection is So, implementing it in

ASP.NET is SIMPLE: Plug-n-Play


4/18

IS 'PCA BASED EIGENFACES' METHOD 100%

EFFICIENT???

NO, and i mean NO face RECOGNITION algorithm is

YET100% efficient ! yes, it could reach 100%

efficiency but NOTALWAYS! So, NO EXISTING face

recognition algorithm is 100% foolproof.

PCA based Eigenfaces method is NOT 100% efficient!

in fact, on the average, it goes up to 70% to 75%

efficiency honestly.


5/18


6/18


7/18


8/18


9/18

Is this a face or not?


10/18

Sketch of a Pattern Recognition

Architecture

Feature

ExtractionClassification

Image

(window)

Object

IdentityFeature

Vector


11/18

Face Detection Algorithm

Face Localization

Lighting Compensation

Skin Color Detection

Color Space Transformation

Variance-based Segmentation

Connected Component &Grouping

Face Boundary DetectionVerifying/ WeightingEyes-Mouth Triangles

Eye/ Mouth Detection

Facial Feature Detection

Input Image

Output Image


12/18

Viola-Jones Face Detection

Algorithm Overview :

Viola Jones technique overview Features

Integral Images

Feature Extraction

Weak Classifiers

Boosting and classifier evaluation

Cascade of boosted classifiers

Example Results

Full details available in paper by Viola and Jones, International Journal of ComputerVision, 2004

Some of the following slides were adapted from a presentation by Nathan Faggian


13/18

Viola Jones Technique Overview Three major contributions/phases of the algorithm :

Feature extraction Classification using boosting

Multi-scale detection algorithm

Feature extraction and feature evaluation.

Rectangular features are used, with a new image representation their

calculation is very fast.

Classifier training and feature selection using a slight variation of a

method called AdaBoost.

A combination of simple classifiers is very effective


14/18

Features

Four basic types.

They are easy to calculate.

The white areas are subtracted from the black ones.

A special representation of the sample called the integral image makes

feature extraction faster.


15/18

Feature Extraction

Features are extracted from sub windows of a sample image.

The base size for a sub window is 24 by 24 pixels.

Each of the four feature types are scaled and shifted across all possible

combinations

In a 24 pixel by 24 pixel sub window there are

~160,000 possible features to be calculated.


16/18

Practical implementation

Details discussed in Viola-Jones paper

Training time = weeks (with 5k faces and 9.5k non-faces)

Final detector has 38 layers in the cascade, 6060features

700 Mhz processor: Can process a 384 x 288 image in 0.067 seconds (in 2003

when paper was written)


17/18

Comparison of classification accuracy

*


18/18

Recognition with Eigenfaces

Algorithm

1. Process the image database (set of images with labels) Run PCAcompute eigenfaces Calculate the K coefficients for each image

2. Given a new image (to be recognized) x, calculate K coefficients

3. Detect if x is a face

4. If it is a face, who is it?

Find closest labeled face in database

nearest-neighbor in K-dimensional space

Documents

Principal Component Analysis Using Eigenfaces for Face Recognition