Handbook of Face Recognition
Editors: Stan Z. Li and Anil K. Jain
Springer. New York. ISBN# 0-387-40595-x.
16 Chapters. 400 pages. Hardcover.
To Order from Bookstore: in USD, in GBP, in EUR, in JPY
Table of Contents
1. Introduction (Stan Li, Chinese Academy of Sciences & Anil Jain, Michigan State University)
1.1 Face Recognition Processing
1.2 Analysis in Face Subspaces
1.3 Technical Challenges
1.4 Technical Solutions
1.5 Current Technology Maturity
2. Face Detection (Stan Li, Microsoft Research Asia)
2.1 Appearance and Learning Based Approach
2.2 Preprocessing
2.2.1 Filtering Using Skin Color
2.2.2 Image Normalization
2.2.3 Multi-Gaussian Clustering
2.3 Neural and Kernel Methods
2.4 Boosting Based Methods
2.4.1 Haar-Like Features
2.4.2 Learning Feature Selection
2.4.3 Learning Weak Classifiers
2.4.4 Boosted Strong Classifier
2.4.5 FloatBoost Learning
2.4.6 Cascade of Strong Classifiers
2.5 Dealing with Head Rotations
2.6 Post-Processing
2.7 Performance Evaluation
3. Modeling Facial Shape and Appearance (Tim Cootes & Chris Taylor, Haizhuang Kang & Vladimir Petrovic, University of Manchester)
3.1 Background
3.2 Statistical Models of Appearance
3.2.1 Statistical Shape Models
3.2.1.1 Aligning sets of shapes
3.2.1.2 Statistical Models of Variation
3.2.1.3 Examples
3.2.2 Statistical Texture Models
3.2.3 Combined Models of Appearance
3.3 Active Shape Models
3.3.1 Modeling Local Structure
3.3.2 ASM Search Algorithm
3.3.3 Examples of Search
3.4 Active Appearance Models
3.4.1 Quality of Match
3.4.2 Iterative Model Refinement
3.4.3 Examples of AAM Search
3.4.4 Alternative Strategies
3.5 Discussion
4. Parametric Face Modeling and Tracking (Jorgen Ahlberg, Swedish Defense Research Agency & Fadi Dornaika, Linkoping University)
4.1 Introduction
4.2 Previous Work
4.3 Parametric Face Modeling
4.3.1 Facial Parameterizations
4.3.1.1 Facial Action Coding System
4.3.1.2 MPEG-4 Facial Animation
4.3.2 Face Models
4.3.3 Adapting a model to an image
4.4 The Tracking Problem
4.4.1 Appearance-based or Featureless Tracking
4.4.2 Feature-based tracking
4.5 Tracker examples
4.5.1 A feature-based tracker using EKF and SfM (Ström)
4.5.2 Appearance-based trackers (Ahlberg, LaCascia)
4.5.3 Combining appearance and feature-based tracking (Dornaika)
4.6 Discussion
5. Illumination Modeling Illumination Modeling for Face Recognition (Ronen Basri, The Weizmann Institute of Science & David Jacobs, University of Maryland)
5.1 Empirical Motivations for Linear Lighting Models
5.2 Linear lighting models: without shadows
5.3 Attached shadows: non-linear constraints
5.4 Spherical Harmonics for lighting modeling
5.4.1 Linear models of attached shadows
5.4.2 Non-linear constraints
5.5 Specularity and cast shadows
5.6 Reconstruction
6. Facial Skin Color Modeling (Birgitta Martinkauppi & Matti Pietikäinen, University of Oulu)
6.1 Introduction
6.2 Skin Colors in Different Color Spaces
6.3 Skin Color Models
6.4 Skin Color Correction
6.5 Comparison of Different Models
6.6 Discussion
7. Face Recognition in Subspaces (Gregory Shakhnarovich, Massachusetts Institute of Technology， Baback Moghaddam, MERL Cambridge Research)
7.1 Dimensionality
7.1.1 Representation, Oversampling (Nyquist and beyond)
7.1.2 "Imagespace": Intrinsic Degrees-of-Freedom
7.1.3 "Facespace": Pixels vs. Basis Functions
7.1.4 Linear Algebraic Concepts
7.1.4.1 KLT, SVD and Eigenvector Decompositions
7.1.4.2 Rank, DOF and the Eigenspectrum
7.2 Linear Subspaces
7.2.1 Neural Network formulations
7.2.2 Karhunen-Loeve, PCA and "Eigenfaces"
7.2.3 Linear Discriminants: "Fisherfaces"
7.2.4 "Dual Eigenspace" (Bayesian) Methods
7.2.5 Factor Analysis, ICA & Source Separation
7.2.6 Multidimensional SVD: "Tensorfaces"
7.2.7 Local Feature Analysis [optional]
7.3 Nonlinear Subspaces
7.3.1 Auto-Encoder MLPs
7.3.2 Principal Curves/Surfaces (NLPCA & Regression)
7.3.3 Kernel-PCA and Kernel-Fisher Methods
7.3.4 IsoMap, LLE and variants [optional]
7.4 Methodology and Usage [brief discussion + pointers]
7.4.1 Appearance-Based Representations
7.4.2 Multiple View-Based Approach for Pose
7.4.3 Fisher & "Illumination Cones" for Lighting
7.4.4 2D/3D Shape-Texture Models
8. Face Tracking and Recognition from Video (Rama Chellappa, ShaoHua Zhou, University of Maryland)
8.1 Review
8.2 Simultaneous Tracking and Recognition from Video
8.2.1 A Time Series State Space Model for Recognition
8.2.2 The Posterior Probability of Identity Variable
8.2.3 Sequential Importance Sampling Algorithm
8.2.4 Experimental Results
8.3 Enhancing Tracking and Recognition Accuracy
8.3.1 Modeling Inter-frame Appearance Changes
8.3.2 Modeling Appearance Changes between Frames and Gallery Images
8.3.3 Experimental Results
8.4 Issues and Discussions References
9. Face Recognition across Pose and Illumination (Ralph Gross, Simon Baker, Iain Matthews, Takeo Kanade, Carnegie Mellon University)
9.1 Review
9.2 The CMU Pose, Illumination, and Expression (PIE) Database
9.3 Eigen Light-Fields for Face Recognition Across Pose
9.4 Normalizing for Illumination
9.5 Modeling Pose and Illumination
10. Morphable Models of Faces (Sami Romdhani, University of Basel & Volker Blanz, Max-Planck-Institut fur Informatik & Curzio Basso & Thomas Vetter, University of Basel)
10.1 Morphable Model for Face Analysis
10.1.1 Three dimensional representation
10.1.2 Correspondence based representation
10.1.3 Face Statistics
10.2 3D Morphable Model Construction
10.2.1 Dense correspondences computed by optical flow
10.2.2 Face Space
10.3 A Morphable Model to Synthesize Images
10.3.1 Shape Projection
10.3.2 Inverse Shape Projection
10.3.3 Illumination and Color Transformation
10.3.4 Image Synthesis
10.4 Image Analysis with a 3D Morphable Model
10.4.1 Stochastic Newton Descend
10.4.2 Inverse Image Compositional Alignment
10.5 Identification
10.5.1 Face Images Databases
10.5.2 Pose Variation
10.5.3 Pose and Illumination Variations
10.5.4 Identification Confidence depends on Fitting Accuracy
10.5.5 Virtual views as an aid to standard face recognition algorithms
11. Facial Expression Analysis (Ying-li Tian, IBM Watson Research Center & Takeo Kanade, Carnegie Mellon University & Jeffrey Cohn, University of Pittsburgh)
11.1 Introduction
11.2 Problem Space for Face Expression Analysis
11.2.1 Level of Description
11.2.2 Individual Differences in Subjects
11.2.3 Degrees of Facial Expression
11.2.4 Databases
11.2.5 Reliability of Ground truth
11.2.6 Lighting Changes
11.2.7 Head Orientation and Scene Complexity
11.2.8 Relation to Other Facial Behavior or Non-facial Behavior
11.3 Automatic Facial Expression Analysis
11.3.1 Face Detection or Head Detection
11.3.2 Facial Data Extraction
11.3.3 Feature Based Methods
11.3.4 Template Based Methods
11.3.5 Facial Expression Recognition
11.3.6 Facial Expression Recognition from Static Image
11.3.7 Facial Expression Recognition from Image Sequences
11.4 Discussion
12. Face Synthesis (ZiCheng Liu, Bai-Ning Guo, Microsoft Research)
12.1 Review
12.2 Face Modeling
12.2.1 Face modeling from an image sequence
12.2.2 Face modeling from two views
12.2.3 Face modeling from a single view
12.3 Face relighting
12.4 Facial animation capturing
12.5 Speech driven facial animation
12.6 Facial expression mapping
12.7 Expression morphing
12.8 Expression detail synthesis
13. Face Databases (Ralph Gross, Carnegie Mellon University)
13.1 Databases for Face Detection
13.1.1 MIT-CMU Test Set
13.1.2 CMU Test Set II
13.1.3 Others
13.2 Databases for Face Recognition
13.2.1 FERET
13.2.2 Yale Face Database B
13.2.3 AR Database
13.2.4 PIE
13.2.5 Notre Dame HumanID database
13.2.6 Asian Face Database
13.2.7 Others
13.3 Databases for Expression Analysis
13.3.1 JAFFE Dataset
13.3.2 Cohn-Kanade DB
13.3.3 Others
13.4 Other Modalities
13.4.1 3D Faces
13.4.2 Hyperspectral Images
14. Performance Evaluation (P. Jonathan Phillips, Patrick Grother, Ross Micheals, National Institute of Standards and Technology)
14.1 Introduction
14.2 Performance Measures
14.2.1 Watch List
14.2.2 Verification
14.2.3 Identification
14.3 FRVT 2002 Protocol
14.3.1 Similarity Scores
14.3.2 Virtual Galleries and Probe Sets
14.3.3 Normalization
14.4 Variability and Demographics
14.5 Advanced Statistical Techniques
14.6 Conclusion
15. Psychological and Neural Perspectives (Alice J. O'Toole, University of Texas)
15.1 Characteristics of Human Face Recognition
15.1.1 Extracting information from the human face
15.1.1.1 Identity
15.1.1.2 Visual categories: sex, age, race
15.1.1.3 Movement and social signals
15.1.1.4 Facial expressions and emotions
15.1.2 Factors that affect performance
15.1.2.1 Stimulus factors
15.1.2.2 Subject factors
15.1.2.3 Photometric factors
15.2 The Neural Systems Underlying Human Face Recognition
15.2.1 The Multiple Systems Model
15.2.1.1 Fusiform face area: identification and categorization
15.2.1.2 Superior temporal sulcus - Movement and social signals
15.2.1.3 Neural processing of emotion from facial expression
16. Face Recognition Applications (Thomas Huang, Ziyou Xiong, ZhenQiu Zhang, University of Illinois)
16.1 Introduction
16.2 Face ID
16.3 Access Control
16.4 Security
16.5 Surveillance
16.6 Smart Cards
16.7 Law Enforcement
16.8 Face Databases Applications
16.9 Multimedia Management
16.10 Human Computer Interaction
16.11 Other Applications
16.12 Limitations of Current Face Recognition Applications