Human Recognition in Unconstrained Environments: Using Computer Vision, Pattern Recognition and Machine Learning Methods for Biometrics

Front Cover
Human Recognition in Unconstrained Environments
Copyright
Contents
Contributors
Editor Biographies
Foreword
1 Unconstrained Data Acquisition Frameworks and Protocols
	1.1 Introduction
	1.2 Unconstrained Biometric Data Acquisition Modalities
	1.3 Typical Challenges
		1.3.1 Optical Constraints
		1.3.2 Non-comprehensive View of the Scene
		1.3.3 Out-of-Focus
		1.3.4 Calibration of Multi-camera Systems
	1.4 Unconstrained Biometric Data Acquisition Systems
		1.4.1 Low Resolutions Systems
		1.4.2 PTZ-Based Systems
		1.4.3 Face
	1.5 Conclusions
	References
2 Face Recognition Using an Outdoor Camera Network
	2.1 Introduction
	2.2 Taxonomy of Camera Networks
		2.2.1 Static Camera Networks
		2.2.2 Active Camera Networks
		2.2.3 Characteristics of Camera Networks
	2.3 Face Association in Camera Networks
		2.3.1 Face-to-Face Association
		2.3.2 Face-to-Person Association
	2.4 Face Recognition in Outdoor Environment
		2.4.1 Robust Descriptors for Face Recognition
		2.4.2 Video-Based Face Recognition
		2.4.3 Multi-view and 3D Face Recognition
		2.4.4 Face Recognition with Context Information
		2.4.5 Incremental Learning of Face Recognition
	2.5 Outdoor Camera Systems
		2.5.1 Static Camera Approach
		2.5.2 Single PTZ Camera Approach
		2.5.3 Master and Slave Camera Approach
		2.5.4 Distributed Active Camera Networks
	2.6 Remaining Challenges and Emerging Techniques
	2.7 Conclusions
	References
3 Real Time 3D Face-Ear Recognition on Mobile Devices: New Scenarios for 3D Biometrics \"in-the-Wild\"
	3.1 Introduction
	3.2 3D Capture of Face and Ear: CURRENT Methods and Suitable Options
		3.2.1 Laser Scanners
		3.2.2 Structured Light Scanners
		3.2.3 Stereophotogrammetry
	3.3 Mobile Devices for Ubiquitous Face-Ear Recognition
	3.4 The Next Step: Mobile Devices for 3D Sensing Aiming at 3D Biometric Applications
	3.5 Conclusions and Future Scenarios
	References
4 A Multiscale Sequential Fusion Approach for Handling Pupil Dilation in Iris Recognition
	4.1 Introduction
		4.1.1 Pupil Dilation
		4.1.2 Layout
	4.2 Previous Work
		4.2.1 Pupil Dilation
		4.2.2 Bit Matching
	4.3 WVU Pupil Light Reflex (PLR) Dataset
	4.4 Impact of Pupil Dilation
	4.5 Proposed Method
		4.5.1 IrisCode Generation
		4.5.2 Typical IrisCode Matcher
		4.5.3 Multi-filter Matching Patterns
		4.5.4 Proposed IrisCode Matcher
	4.6 Experimental Results
	4.7 Conclusions and Future Work
	References
5 Iris Recognition on Mobile Devices Using Near-Infrared Images
	5.1 Introduction
	5.2 Preprocessing
	5.3 Feature Analysis
	5.4 Multimodal Biometrics
	5.5 Conclusions
	References
6 Fingerphoto Authentication Using Smartphone Camera Captured Under Varying Environmental Conditions
	6.1 Introduction
	6.2 Literature Survey
	6.3 IIITD SmartPhone Fingerphoto Database v1
		6.3.1 Set 1: Background Variation
		6.3.2 Set 2: Illumination Variation
		6.3.3 Set 3: Live-Scan Fingerprints
	6.4 Proposed Fingerphoto Matching Algorithm
		6.4.1 Fingerphoto Segmentation
		6.4.2 Fingerphoto Enhancement (Enh#1)
		6.4.3 LBP Based Enhancement (Enh#2)
		6.4.4 Scattering Network Based Feature Representation
		6.4.5 Matching Techniques
	6.5 Experimental Results
		6.5.1 Performance of the Proposed Matching Pipeline
		6.5.2 Comparison of Matching Algorithms
		6.5.3 Comparison of Distance Metrics
		6.5.4 Effect of Enhancement
	6.6 Conclusion
	6.7 Future Work
	Acknowledgements
	References
7 Soft Biometric Attributes in the Wild: Case Study on Gender Classification
	7.1 Introduction
	7.2 Biometrics in the Wild
	7.3 Gender Classification in the Wild
		7.3.1 Datasets
		7.3.2 Proposals Summary
		7.3.3 Discussion
	7.4 Conclusions
	References
8 Gait Recognition: The Wearable Solution
	8.1 Machine Vision Approach
	8.2 Floor Sensor Approach
	8.3 Wearable Sensor Approach
		8.3.1 The Accelerometer Sensor
	8.4 Datasets Available for Experiments
	8.5 An Example of a Complete System for Gait Recognition
	8.6 Conclusions
	References
9 Biometric Authentication to Access Controlled Areas Through Eye Tracking
	9.1 Introduction
	9.2 ATM-Like Solutions
	9.3 Methods Based on Fixation and Scanpath Analysis
	9.4 Methods Based on Eye/Gaze Velocity
	9.5 Methods Based on Pupil Size
	9.6 Methods Based on Oculomotor Features
	9.7 Methods Based on Head Orientation
	9.8 Conclusions
	References
10 Noncooperative Biometrics: Cross-Jurisdictional Concerns
	10.1 Introduction
	10.2 Biometrics for Implementing Biometric Surveillance
	10.3 Reaction to Public Opinion
		10.3.1 Geopolitical Context
		10.3.2 Technological Skills
		10.3.3 Proportionality
		10.3.4 A Particular Operational Framework
	10.4 The Early Days
		10.4.1 Commercial Context
		10.4.2 Historical Context
		10.4.3 Social Context, the Newham and Ybor City Experiments
	10.5 An Interesting Clue (2007)
	10.6 Biometric Surveillance Today
		10.6.1 Increased Perception of Insecurity
		10.6.2 Getting Used to the Erosion of Privacy
		10.6.3 Increase of Mobility
	10.7 Conclusions
	References
Index
Back Cover