CS 7321: Computer Vision (Winter 1998)

CS 7322: Computer Vision II

(aka. High-Level Computer Vision)
Information/Administrivia

Spring 1998 4:35pm-5:55pm, April 1, 1998 - June 5, 1998
Classroom: CoC 102 (Monday, Wednesday)
Instructor: Irfan Essa

Instructor

Name:	Irfan A. Essa
Email:	irfan@cc.gatech.edu	PREFERRED FORM OF COMMUNICATION *(put CS7321 in subject line)*
Office:	320 CoC	404-894-6856

What is Computer Vision?

Computer vision, sometimes called image understanding, describes the automatic deduction of the structure and the properties of a possibly dynamic three-dimensional world from either a single or multiple two-dimensional images of the world. The images maybe monochromatic or colored, they may be captured by a single camera or multiple cameras, and each camera may be either stationary or mobile.

Description of Course

An advanced course in computer vision and machine perception. This quarter we will study different methods for motion-based recognition of human activity from video.

Topics to be covered

Analysis of Image Sequeces
Image Motion
Recognition from Image Motion
Human Activities

whole body
faces

Kalman Filtering
HMMs

Prerequisites

Linear Algebra Some Probability Some Programming Knowledge	OR	CS7321	OR	Permission of Instructor

Text(s):

Required text:
None.

Recommended Texts (not required, just suggested for additional material)

Motion-based recognition, M. Shah and R. Jain (Editors), Kluwer-Academic Publishers, 1997. (WOULD HAVE BEEN REQUIRED, except it was too expensive).
Machine Vision R. Jain, E, Kasturi, and B. G, Schunck, McGraw-Hill, 1995.
Foundations of Vision, B. A. Wandell, Sinauer Associates Inc. 1995,
High-level Vision: Object Recognition and Visual Cognition , S, Ullman, MIT Press 1995.
Computer and Robot, Vision Volume I and II, R. M. Haralick and L. G. Shapiro, Addison-Wesley, 1993.
A guided tour of Computer Vision, V. S. Nalwa, Addison-Wesley 1993.

Some of the readings from these texts will be handed out in class.

Additional Reading(s):

Additional Readings in form of state-of-the-art papers or chapters from other text books will be handed out in class.

WWW/Email

Most of the class will extensively use the WWW. The schedule and the lecture notes will be made available off the CLASS HOME page. All reports and student submissions are required in WEB readable format. Further information on that will be provided.

Email will be the primary form of communication after the class period. Please make sure to inform the instructor of your permenant email address.

Grading

Commentary and Critique on Assigned Reading (One Page)

Due electronically by 10am of the day of the class the reading was assigned for, 24% of the Grade.

8 commentaries will be assigned. 2 will be optional

Leading a class discussion and presenting an assigned paper, 20% of the grade
Midterm Project (OR Progress Report) 10% of the grade.
A Final Project 36% of the grade. (TOTAL for final Project: 46%)
Class Participation, 10% of the grade.

Final Projects

More to come.

Policies

Honor Code:: Academic Honor Code instituted by the Students of Georgia Tech applies in this class.
Collaboration:: The critique and summary of readings HAVE TO BE DONE ON YOUR OWN WITHOUT ANY COLLABORATION.; Discussions on projects and problem sets is allowed, however, the solutions to the problems sets and projects have to be done individually. In case of discussions in groups, please specify to me who were the members of the group that came to the submitted solutions.
Reports:: All written material will be submitted electronically in HTML format and will be put up on the WWW pages for this course. If for some reason you object to this please see the instructor.

Office Hours

By appointment via email

Computing Resources

We will use the following two labs for work in this class:

SGI Indy Cluster (130 CoC)

Hours Posted on the Door

Imaging Lab (304 CoC)

M-Th 9a-10p, F 9a-6p, S-Su 12n-6p

Newsgroup

There is a newsgroup for this class. git.cc.class.7322. Be warned that I am not much of newsgroup reader.

Back to

CS7321 Home Page,

College of Computing Home Page

(aka. High-Level Computer Vision) Information/Administrivia