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 3dface - End User Meeting 2

Multibiometrics Biometric Biometric Biometric Biometric Score Biometric Algorithm Sample Score Identifier Algorithm Sample Biometric Biometric Biometric Biometric Score Biometric Sample Algorithm Identifier Score Algorithm Sample Biometric Score Algorithm • Multiple biometric identifier •Multiple biometric samples Fusion Engine •at different sample qualities •Multiple biometric algorithms •One combined score Score 27.03.2007 3dface - End User Meeting 3

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

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 October 23, 2006 FaceVACS Algorithms: An Overview Page 5

Multibiometrics using the face Intensity Intensity Score Data Intensity Algorithm data Skin Skin Texture Texture Score Algorithm Data Shape Shape Shape Score Algorithm Data data Combined Score Algorithm Alligned biometric samples Fusion •Intensity data Engine •Shape data Scor e 27.03.2007 3dface - End User Meeting 6

Sample FRGC images Controlled Uncontrolled July 2006 http://www.cognitec.com 7

Sample frontal Yale images Varying lighting conditions July 2006 http://www.cognitec.com 8

Example: shape data Monocular, fixed view vertex data is sensors produce subject to noise occlusions July 2006 http://www.cognitec.com 9

Example: shape data Gap Filling / some vertices plane surface patches have large deviations 27.03.2007 3dface - End User Meeting 10

Example: skin texture Skin texture ananlysis indicates the degree to which two surfaces are the same if the blocks match in an orderly fashion 27.03.2007 3dface - End User Meeting 11

Simultanious Capture Combined Sensor Intensity Intensity identifier Data and High Resolution Skin Texture Camera Skin sample Texture identifier 3d Shape Shape Data Shape identifier Sensor sample Real Person with Digital representation Facial characteristics of biometric samples 27.03.2007 3dface - End User Meeting 12

Fusion Architecture for Face Comparison Engine Intensity Images score Shape Estimator Comparison Engine Estimates shape from single or multiple Intensity and Shape Images intensity images Compensate intensity image using parameters of the shape score Comparison Engine Shape Images Score Fusion Engine Use dependency between the scores to produce a fused one final score July 2006 http://www.cognitec.com 13

Promising Results Equal Error Rates •Using Intensity Images („2d“): 3.14% •Using Shape Data („3d“) 2.54% •Using both (2d+3d): 1.01% 14

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