EE392J: Digital Video Processing

Course Overview

This graduate-level course covers the fundamentals of digital video processing. Coverage includes spatio-temporal sampling, motion analysis, parametric motion models, motion-compensated filtering, and video processing operations including noise reduction, restoration, superresolution, deinterlacing and video sampling structure conversion, and compression (frame-based and object-based methods). A number of advanced topics will be covered including video segmentation and layered video representations, video watermarking, compressed-domain video processing, and digital television.  By the end of the quarter, students will be able to design digital video processing algorithms and gain hands-on experience through the final project.  You can look at the student EE392J final projects from Winter 2002 and Winter 2000, as illustrative examples of the final projects that you can do in this course.

Instructors

John Apostolopoulos, Ph.D. 
     Email: japos@hpl.hp.com
     Phone: 857-4416
     Office Hours: Mon/Wed after class, Tues or Fri (TBA) 4:00-5:00pm, Packard Building 2nd Floor Lounge
Susie Wee, Ph.D. 
     Email: swee@hpl.hp.com
     Phone: 857-7585
     Office Hours: After class

Teaching Assistant

Ana Bertran Ortiz
    Email:  nuska@stanford.edu
    Office:  TBA
    Office hours: TBA

Lectures

Winter quarter, Monday and Wednesday, 4:15-5:30pm, in Redwood Hall, Room G19.  This classroom is located roughly here (Redwood Hall is located slightly to the left of center on this map).

Course Outline

An overview of the topics covered is given in the Course Outline. There may also be one or two invited lectures on emerging topics in digital video processing given by experts on those topics.  The Course Schedule lists the topics of each lecture.

Prerequisites

This course is intended for graduate students with a strong background in digital signal processing. The recommended prerequisites are to have taken at least one of the following: Digital Image Processing (EE368), Image Communication I or II (EE398A or EE398B), or Advanced Topics in Image, Video, and Multimedia Systems (EE368C), or prior experience with image or video processing or computer vision.  If you do not have any of these prerequisites, then permission of the instructors is required.

Problem Sets

Problem sets will be issued approximately once a week, with smaller problem sets toward the end of the quarter when students will be working on their final projects. A few of the problem sets will involve Matlab-based development and simulation of various video processing algorithms.

Problem sets and class handouts will be distributed in class.  Extra copies can be picked up after class from a file cabinet near Packard Room 259 (or from course secretary Kelly Yilmaz in Packard Room 259).   Problem sets can also be accessed here.

Final Project

An important element of this course is the final project where the students, working individually or as part of a team, will work on a problem in digital video processing. A list of possible final projects, as well as other relevant information, is given in the Projects webpage.  You can also look at the final projects from Winter 2002 and Winter 2000, as illustrative examples of the final projects that you can do in this course.
This year's final projects can be found in Winter 2004

Grading (tentative)

25% Problem Sets (exercises and lab assignments)
35% Midterm Exam
40% Final Project

Textbook

The required text for the class is "Video Processing and Communications" by Yao Wang, Joern Ostermann, and Ya-Qin Zhang, Prentice Hall, 2002, ISBN 0-13-017547-1.  The text is available at the Stanford Bookstore.

A recommended text is "Digital Video Processing" by M. Tekalp (Prentice Hall, 1995, ISBN 0-13-190075-7).  A list of corrections for the recommended text is available.

In addition to the material covered in the text,  a number of emerging topics will be discussed that are not covered in the text. Class attendance is highly recommended.

          Additional References

Prior years' Digital Video Processing (EE392J) course webpages: Winter 2002 and Winter 2000