This is a tentative schedule and subject to change.

Date Lecture Description Readings Assignments Materials
9/3 Introduction.  Class ends early (11:30) for commencement ceremonies.     Lecture 1 slides
9/5 Introduction continues.

What is vision good for? Why is it hard?  Why is it interesting? How do you pose the problem computationally?

    Assignment 0 out Lecture 2 slides

Ball and Shadow movie

Illusory motion from shadows

9/8 Continuing Introduction.

Case study - object recognition.

Ch 1.1, 1.4, Ch 4.   Lecture 3 slides
9/10

Convolution and linear filtering.

Ch 7.4-7.7, 9.2

Lecture 4 slides

Matlab code - Gaussian smoothing

9/12 Pyramids and image derivatives   Assignment 0 due

Assignment 1 out

Lecture 5 slides

Matlab code - edges

Introduction to Matlab
Matlab tutorial code

9/15 First derivatives and edges.     Lecture 6 slides

Matlab code - derivatives of Gaussians

Change blindness
  * Flicker
  * Gradual changes
  * Person change 1
  * Person change 2

9/17

Gradients and Laplacian pyramid.

Linear algebra tutorial. CIT 165 (Motorola), 6:30PM - 7:30PM

 


Lecture 7 slides

Linear Algebra Review Slides

9/19 Filtering and features

Linear algebra tutorial. CIT 165 (Motorola), 6:30PM - 7:30PM

Extra reading: feature detection: Distinctive Image Features from Scale-Invariant Keypoints (read Sections 1-3.1)

Extra readings related to the next few lectures

Asgn1 - Problem 1 and 2 Due
Hand-in name: asgn1_p1_p2
Lecture 8 slides
9/22 Images as vectors.  Appearance-based models Ch 22.3
Lecture 9 slides
9/24 Covariance and PCA Extra Chapter
Lecture 10 slides
9/26 PCA and SVD   Asgn1 - Problem 3 and 4 Due
Hand-in name: asgn1all



Lecture 11 slides
9/29 Finish PCA applications.  Ch 22.1, 22.2 Assignment 2 p1 p2 out Lecture 12 slides
10/1 Review of basic probability.Multivariate Gaussians, covariance, probability.     Lecture 13 slides
10/3 Guest Lecture: Real application of PCA.  Modeling human body shape.  Alex Balan Ch 15 Asgn2 Problem 1 due
Hand-in name: asgn2_p1
 
10/6 Finish multivariate Gaussians and PCA   Asgn 2 (problem 3) out Lecture 14 slides
10/8 Motion Intro Moghaddam & Pentland Asgn2 Problem 2 due
Hand-in name: asgn2_p2
 Lecture 15 slides
10/10

Guest Lecture: Joe Mundy

class notes on motion  

 

10/13 University Holiday, No class  
 
10/15

Guest Lecture: Moldovan, Denoising old movies.

Moldovan et al. Assign2 Problem 3 due

 Hand-in name: asgn2all

 
10/17 Guest Lecture: Gabriel Taubin
 
 
10/20

Affine Motion

  Assignment 3 out Lecture 16 slides
10/22 Finish affine motion
    Lecture 17 slides
10/24 Cameras and projection  
Lecture 18 slides
10/27

Robust estimation

Reading: Robust statistics and optical flow  

Assignment 3, Problems 1 & 2 due

Hand-in name: asgn3_p1_p2

Lecture 19 slides
10/29 Robust estimation II, Non-linear optimization  
Lecture 20 slides 
10/31

Guest Lecture: Moldovan, Super Resolution

Reading: Super Resolution, Irani and Peleg



Super Resolution slides
11/3

Robust regularization

Reading: Horn and Schunck, Determining Optical Flow

Assignment 3, All problems due

Hand-in name: asgn3all 

 

Lecture 21 slides
11/5 Dense optical flow

Start tracking


Assignment 4 out

 Good overview article on sampling and particle filters

Particle Filter

Lecture 22 slides

11/7 Tracking intro     
Lecture  23 slides  
11/10 Particle filtering  

Assignment 4, Problem 1due
Hand-in name: asgn4_p1

Lecture 24 slides

Mars Stereo slides

11/12 Finish particle filtering.

More project ideas

  Project handout
 

Lecture 25 slides

 

11/14 Project ideas and Binocular Stereo   Bring project ideas to class Project Ideas

More project info and some links

11/17

Guest Lecture: Bill Warren

 

Assignment 4, all problems due
Hand-in name: asgn4_all  

 
11/19 Stereo   Project proposals Due
Hand-in name: proposal
Lecture 26 slides
11/21 Fields of Experts Fields of Experts   FoE slides part 1 and 2
11/24 Fields of Experts      
11/26 Thanksgiving recess. No Class  
 
11/28 Thanksgiving recess. No Class      
12/1 Learning optical flow      
12/3 Modern object recognition      
12/5 The future      
12/8 Reading week  No class      
12/10 Reading week.  No class  
 
12/12 exam period (no exam)  Project due. Part based recognition: Fergus, Perona & Zisserman

Viola & Jones face detector

Robert Shapire's boosting page

 

Projects due.
Hand-in name: proj