Fall 2007

Tuesday 12.35-13.55, Thursday 9.35-10.55

Professor: Frank Dellaert


At GTL, the usual pre-requisites will be a bit relaxed to make the course more accessible to non-CS students. Familiarity with linear algebra and statistics will be helpful, but not essential. Similarly, familiarity with programming will help, but all projects will be in MATLAB, so this should not be too demanding.

Office Hours and Contact information

Email: dellaert@cc.gatech.edu, please use “CSx495” in the subject line (automatic by clicking the link)
Office hours: Tuesday & Wednesday 2-3pm, Room 206

Class Goals

The desired learning outcomes for the students are:

- know what computer vision is

- knowledge of vision fundamentals

- learn a methodology to attack vision problems

- see examples of vision problems and their solutions

- gain familiarity with popular computational techniques


I will not be using a required textbook this year, but there are two texts that I will use as inspiration:

- Computer Vision: A Modern Approach. David Forsyth and Jean Ponce. Prentice Hall.

- Multiple View Geometry in Computer Vision, by Richard Hartley and Andrew Zisserman. Cambridge University Press.


I will be using powerpoint slides in combination with the blackboard. The powerpoint presentations will be available before class so you can print them out and take notes, if desired.


There will be 4 programming assignments (in MATLAB) and one larger project in the middle of the semester (due date right before fall break). The project can be collaborative and will be a project of your choosing, either from a list of suggestions provided by the instructor

We will use MATLAB for the projects, or Processing for those that want it. To get familiar with MATLAB, please consult the following links:

- "MATLAB Primer", Kermit Sigmon

- "A Practical Introduction to Matlab", Mark Gockenbach

- "Official Matlab Usenet FAQ"

- "The Matlab Manuals", Mathworks

In addition to the projects, there will be a small weekly exercise to prime your thought about the topic of the next week. They are graded at about 1% each and are not expected to be much work at all.

Late policy: 1 day late: 50% of the grade, 2 days late: 25% of the grade, later than that: 0% of the grade. It is always to ask prior approval to hand in an assignment late because of special reasons.

Collaboration Policy

Collaboration on assignments is encouraged at the "white board interaction" level. That is, share ideas and technical conversation, but write your own code. Students are expected to abide by the Georgia Tech Honor Code. Honest and ethical behavior is expected at all times. All incidents of suspected dishonesty will be reported to and handled by the office of student affairs.


Class Attendance & Participation: 5%
Exercises: 15%
Assignments: 40%

Project: 20%
Midterm & Final Exam: 10% each

Graduate Section

To receive graduate credit for the class requires a more complete write-up of the project, which will be due one week later than for the undergraduate students in the class.  I will expect the write-up to be in the style of a scientific conference paper. One thing this will require is a good review of the literature.