Multibiometrics for Face Recognition 3D Face Project End User - - PowerPoint PPT Presentation

Multibiometrics for Face Recognition

3D Face Project – End User Meeting 2007-03-22 / Darmstadt

Volker Kempert (Cognitec Systems GmbH)

Agenda

• Multibiometrics in general
• Multibiometrics related to the face
• Biometric Face Identifiers
• Capturing Biometric Face Identifiers
• Fusion related to the face
• Conclusions

27.03.2007 2 3dface - End User Meeting

Multibiometrics

27.03.2007 3dface - End User Meeting 3

Biometric Algorithm Biometric Algorithm Biometric Algorithm Biometric Algorithm Fusion Engine Biometric Sample Biometric Sample Biometric Sample Biometric Sample Biometric Algorithm

Score Score Score Score Score Score

Biometric Identifier Biometric Identifier

• Multiple biometric identifier
• Multiple biometric samples
• at different sample qualities
• Multiple biometric algorithms
• One combined score

Why Multibiometrics?

• Compared to single biometrics identifier

– Higher Accuracy More secure – More robust – Higher fraud resistance

• Disadvantages

– More complex biometric capturing processes – More complex devices and algorithms Higher operational costs

27.03.2007 3dface - End User Meeting 4

October 23, 2006 FaceVACS Algorithms: An Overview Page 5

Multi-Modal Biometrics

Example for worse performance (John Daugman, 1999):

• Biometric system A: EER = 1%
• Biometric system B: EER = 0.1%

Have A and B operate at their EERs; conduct 100,000 verification attempts with impostors, 100,000 with authentics; then:

• A alone: 2000 errors; B alone: 200 errors
• “AND” rule: 1099 FR’s, 1 FA 1100 errors
• “OR” rule: 1099 FA’s, 1 FR 1100 errors

Multibiometrics using the face

27.03.2007 3dface - End User Meeting 6

Intensity Algorithm Shape Algorithm Fusion Engine Intensity data Shape data

Combined Algorithm

Score Score Score Score Scor e Intensity Data Shape Data Skin Texture Data

Skin Texture Algorithm

Alligned biometric samples

• Intensity data
• Shape data

July 2006 http://www.cognitec.com 7

Sample FRGC images

Controlled Uncontrolled

July 2006 http://www.cognitec.com 8

Sample frontal Yale images

Varying lighting conditions

July 2006 http://www.cognitec.com 9

Example: shape data

vertex data is subject to noise Monocular, fixed view sensors produce

• cclusions

Example: shape data

27.03.2007 3dface - End User Meeting 10

Gap Filling / plane surface patches some vertices have large deviations

Example: skin texture

27.03.2007 3dface - End User Meeting 11

Skin texture ananlysis indicates the degree to which two surfaces are the same if the blocks match in an

• rderly fashion

Simultanious Capture

27.03.2007 3dface - End User Meeting 12

Intensity Data and Skin Texture sample Shape Data sample

Intensity identifier Shape identifier

Skin Texture identifier

Combined Sensor High Resolution Camera 3d Shape Sensor Real Person with Facial characteristics Digital representation

• f biometric samples

July 2006 http://www.cognitec.com 13

Fusion Architecture for Face

Fusion Engine

Use dependency between the scores to produce a fused

• ne

Comparison Engine

Intensity Images

Comparison Engine

Shape Images

Comparison Engine

Intensity and Shape Images Compensate intensity image using parameters of the shape

Shape Estimator

Estimates shape from single or multiple intensity images score score Score final score

14

Promising Results

Equal Error Rates

• Using Intensity

Images („2d“): 3.14%

• Using Shape Data

(„3d“) 2.54%

• Using both (2d+3d):

1.01%

Promising Results (2)

27.03.2007 3dface - End User Meeting 15

Conclusions

• Multi (modal) biometrics help to

improve accuracy compared to single biometrics

• Face is an object that allows the

simultaneous capturing of multi- biometrics identifiers

• Multi-biometrics systems are more

difficult to outsmart

27.03.2007 3dface - End User Meeting 16