Mon/Wed: 12:00-1:20 Winter 2011, Boelter 4413.
This course models vision as Bayesian Inference. It concentrates on visual tasks such as segmenting images, detecting objects in images, and recognizing objects. Its goal is to describe the state of the art techniques. The handouts consist of copies of the lecture notes and related papers.
3 homework assignments (20% each). Term project (40%).
Final
Project Due 20/March. Hand in to Prof. Yuile. Office or Mailbox
| Lecture | Date | Topics | Reading Materials | Handwritten Notes | Handouts |
| 1 | 01-03 | Introduction to the Course: Statistical Edge Detection |
|
Lecture1.pdf | chp1.pdf |
| |
01-05 | No Lecture | |||
| 2 |
01-10 |
Probability Distributions on
Graphs: Basic Introduction |
Lecture2.pdf |
GYtics.pdf |
|
| 3 |
01-12 |
Maximum Likelihood and
Discriminative Training: Without hidden variables. ML versus AdaBoost |
Lecture3.pdf |
||
| 01-17 |
Martin Luther King Holiday |
||||
| 4 |
01-19 |
Maximum Likelihood Learning: Without hidden variables. |
Lecture4.pdf |
chp3.pdf |
|
| 01-24 |
No Lecture |
||||
| 5 |
01-26 |
Entropy and Model Selection |
Lecture5.pdf |
||
| 6 |
01-31 |
Hidden Markov Models |
HMM_example.pdf |
Lecture7.pdf Lecture7.5.pdf |
chp6.pdf |
| 7 |
02-02 |
Free Energy Optimization Methods I Mean Field Theory, Variational Methods |
MeanFieldTheory.PDF |
YuilleMicroBook.pdf |
|
| 8 |
02-04 |
Free Energy Optimization Methods
II Belief Propagation and TRW |
BeliefPropagation.PDF |
||
| 9 |
02-07 |
Altenative Optimzization Algorithms MCMC, Graph Cuts, Linear Programming |
Alternatives.PDF |
||
| 10 |
02-09 |
Max Margin Techniques Structure SVM, Latent SVM, relations to Probabilistic Learning |
ProbabilisticMachineLearning.pdf |
Lecture9.pdf |
yu_joachims_09a.pdf |
| 11 |
02-14 |
Hierarchical Image Labeling Learning a Probabilistic Model for Image Labeling |
imgparsing09pami_d.pdf |
||
| 12 |
02-16 |
Image Parsing Region Competition and Image Parsing |
region_competition_pami.pdf |
ImageParsing.PDF |
|
| 02-21 |
President's Day |
||||
| 02-23 |
Probabilitsic Object Models Probabilistic Models of Objects |
||||
| 02-28 |
Learning Object Models Learning Probabilistic Models of Objects |
||||
| 03-02 |
Learning Hierarchical Object Models Learning Hierarchical Probabilistic Models of Objects |
||||
| 03-07 |
Hierarchical Parts Model and Latent
SVM Pascal Challenge |
Latent3SVM10cvpr.pdf |
|||
| 03-09 |
Latent Boltzmann Machines, Deep Belief Nets, and Active Apperance | deepbelief.pdf |
Lecture14.pdf |