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.
| Lecture | Date | Topics | Reading Materials | Handwritten Notes | Handouts |
| 1 | Day 1 | Introduction |
Vision_and_the_brain |
Intro |
|
| 2 |
Day 1 |
Basic Images |
Images |
||
| 3 |
Day 1 |
Statistical Edge Detection | EdgeDetectionExamples |
Edges |
|
| 4 |
Day 1 |
Piecewise Smoothness |
TotalVariationDenoising |
WeakSmoothness |
|
| 5 |
Day 1 |
Manhattan World |
ManhattanWorld |
Manhattan |
|
| 6 |
Day 2 |
Image Labeling | LabelingRegions |
Labeling |
|
| 7 |
Day 2 |
MarkovRandomFields | TutorialBayes |
MRFs |
|
| 8 |
Day 2 |
InferenceMRFs | Review_MRF_Inference |
Inference |
RevChapter |
| 9 |
Day 2 |
LearningMRFs | GrabCutExamples |
LearnMRF |
FRAME |
| 10 |
Day 2 |
LearningMRFsHidden |
ReviewVisionModels |
LearnHidden |
|
| 11 |
Day 3/4 |
Dynamic Programming and Stereo |
BayesStereo |
DP&Stereo |
|
| 12 |
Day 3/4 |
Review of Learning |
RevLearn |
||
| 13 |
Day 3/4 |
Hidden Markov Models |
HMMexample |
HMMs |
HMM_notes |
| 14 |
Day 3/4 |
Support Vector Machines |
SVMs |
||
| 15 |
Day 3/4 |
AdaBoot and Faces |
ViolaJones |
AdaBoost
ExtraBoost |
|
| 16 |
Day 5 |
Spectral Clustering and
Superpixe;ls |
Superpixels |
SpectralClustering |
|
| 17 |
Day 5 |
Region Competition and Image
Parsing |
region-competition |
RegionCompetition |
|
| 18 |
02-28 |
Lighting Models | GBRambiguity |
Lighting |
SVD |
| 19 |
03-02 |
Active Apperance Models and
FORMS |
AMMsExample |
AAMs |
FORMS |
| 20 |
03-07 |
Deformable Templates |
DTs |
||
| 21 |
03-09 |
Structure Machine Learning/
Hierarchical Models |
RCMs |
SML |