SIGGRAPH 2004 - The 31st international conference on computer graphics and interactive techniques
Conferenece Exhibition Call For Participation Presenters Media Registration spacer
line


21. Introduction to Bayesian Learning
Monday, Tutorial, 3:45 - 5:30 pm
Room 515A
Level: Intermediate

Sophisticated computer graphics applications require complex models of appearance, motion, natural phenomena, and even artistic style. Such models are often difficult or impossible to design by hand. Recent research demonstrates that, instead, we can "learn" a dynamical and/or appearance model from captured data, and then synthesize realistic new data from the model. For example, we can capture the motions of a human actor and then generate new motions as they might be performed by that actor. Bayesian reasoning is a fundamental tool of machine learning and statistics, and it provides powerful tools for solving otherwise-difficult problems of learning about the world from data. Beginning from first principles, this course develops the general methodologies for designing learning algorithms and describes their application to several problems in graphics.

Prerequisites
Familiarity with linear algebra, calculus, and computer graphics.

Intended Audience
Computer graphics researchers and practitioners working on data-driven computer graphics problems, such as animating shape and motion from video or animating from motion capture.

Organizer and Lecturer
Aaron Hertzmann
University of Toronto

Schedule
3:45 Introduction
The Future of Graphics:
  • Data-driven Analysis and Synthesis
  • The Need for Bayesian Reasoning
Hertzmann

4 Fundamentals of Bayesian Probabilistic Reasoning:
  • Classical (Aristotelian) Logic and its Limitations
  • Cox Axioms
  • Bayes Rule
  • Prediction and Parameter Estimation
  • Learning Multinomials and Gaussians
  • Relation to Least-squares Fitting and Frequentist Methods
Hertzmann

4:30 How to Design Learning Algorithms:
  • Generative Models
  • The Overfitting Problem in MAP Learning and Least-squares
  • Marginalization, The EM Algorithm at a Glance
  • Example: Probabilistic PCA
  • Example: Automatic Non-rigid Modeling From Video
Hertzmann

5:15 The Summary and Conclusions:
  • Pros and Cons of the Bayesian Approach
  • Audience Questions
Hertzmann

 
COURSE NUMBERS
1       2       3      
4       5       6      
7       8       9      
10     11     12
13     14     15
16     17     18
19     20     21
22     23     24
25     26     27
28     29     30
31     32     33
Course Evaluation
All course attendees are encouraged to evaluate SIGGRAPH 2004 Courses content and presenters.

Online Evaluation Forms

Your ratings and comments are very important. They will be used to ensure that the annual SIGGRAPH conference consistently offers excellent courses on topics that are important to the SIGGRAPH community.
 > courses committee
REFER A FRIEND
 > share the SIGGRAPH 2004 web site
   
   
 
Conference 8-12 August, Exhibition 10-12 August.  In Los Angeles, CA