CS786-102 - Face Recognition and Video Processing - Spring'2005

• This course focuses on the current state of knowledge of statistical pattern recognition and video processing with application to face detection, tracking, and recognition. Topics include statistical decision theory, classifier design, component analysis, discriminant analysis, the recent statistical learning theory, face detection using neural networks and statistical techniques, face recognition using linear methods, kernel methods, and 3D model-based methods, video processing for face detection, and video processing for face tracking by means of HMM, Kalman filter, and Condensation.

• K. Fukunaga, Introduction to Statistical Pattern Recognition, 2nd edition, Morgan Kaufmann, 1990.
• R.O. Duda, P.E. Hart, and D.G. Stork, Pattern Classification, 2nd edition, John Wiley & Sons, 2001.
• V. N. Vapnik, The Nature of Statistical Learning Theory, 2nd edition, Springer, 2000.
• B. Scholkopf and A. J. Smola,  Learning with Kernels: Support Vector Machines, Regularization, Optimization and Beyond, MIT Press, 2002.
  
• S. Haykin, Neural Networks - A Comprehensive Foundation, 2nd edition, Prentice-Hall, 1999.
• Selected papers.

1. Introduction
• Pattern Recognition Fundamentals
• Face Detection, Tracking, and Recognition
• Video Processing Fundamentals
• Software Tools: IC Imaging, OpenCV, Matlab
  
2. Decision Theory
   
• The Bayes Decision Rule for Minimum Error
• The Bayes Decision Rule for Minimum Cost
3. Classifier Design
• The Bayes Linear Classifier
• Linear Classifier Design
4. Component Analysis
• Linear Component Analysis
• Nonlinear Component Analysis - Kernel Methods
5. Discriminant Analysis
• Linear Discriminant Analysis
• Nonlinear Discriminant Analysis - Kernel Methods
6. Statistical Learning Theory (SLT)
   
• Structural Risk Minimization (SRM)
  
• Support Vector Machines (SVM)
7. Face Detection
   
• Face Detection using Neural Networks
• Face Detection using Statistical Techniques
  
8. Face Recognition - Linear Methods
   
• Eigenfaces and Performance Analysis
  
• Fisherfaces and Generalization Analysis
  
9. Face Recognition - Kernel Methods
   
• Face Recognition using Kernel PCA
  
• Face Recognition using Kernel Discriminant Analysis
  
10. Face Recognition - 3D Model-based Methods
• Shape From Shading and 3D Face Recognition
  
• Range Data Processing and 3D Face Recognition
11. Video Processing and Face Detection
    
• Motion Analysis
  
• Face Detection using Motion Analysis and SVM
  
12. Video Processing and Face Tracking
    
• Face Tracking using HMM
• Face Tracking using Kalman Filter
  
• Face Tracking using Condensation

Projects (topics are related to our course Contents)

Presentation

Paper (~10 pages)

Class attendance

• The Face Recognition Home Page
• Face Detection Home Page
• MATLAB
• Matlab Primer (Third Edition, By Kermit Sigmon - pdf file)
• Matlab Getting Started (copyright Mathworks - pdf file)
• Matlab Image Processing Toolbox User's Guide (copyright Mathworks - pdf file)