DCS 534 - Computer Vision - Spring 2002


Course Information
Description
Schedule
Readings

MATLAB

Course Staff
Doug DeCarlo

Office Hours
Tuesday, 3:00-5:00 pm
CoRE 310

Description

This course aims to provide an understanding of the processes involved in the formation of images of visual scenes, of how computational approaches for transforming, estimating or recognizing such images are formulated and implemented, and of where these methods can and have been applied. The course will also teach implementation and practical use of a wide variety of vision algorithms.

This course is intended for computer science graduate students, as well as students in allied areas (such as psychology or biomedical engineering) who have interests in computational vision and its applications.

Meeting time:

  • Thursdays, 2:50-5:50 pm, Hill 254 (Busch).

Pre-requisite:

Work:

  • Small to medium-sized homeworks (in MATLAB)
  • Course project

Schedule

Date Topics
January 24 Intro: Visual Perception as Computation
January 31 Images
  • S&S: Chapters 1 and 2
  • Marr, Vision: Chapter 1 and 2.1
February 7 Morphology, Color, Segmentation
  • S&S: Chapter 3; Sections 6.1-6.5; Sections 10.1-10.2;
       Chapter 4 (as review of 530)

Homework 1 (due 2/21)

February 14 Filtering, Edge and Feature detection
  • S&S: Chapter 5, Sections 10.3-10.5.
  • Marr, Vision: Section 2.2
Matlab diary (on paul): ~decarlo/534/matlab/lec2-14.m
February 21 Perceptual Organization
  • S. Palmer, Vision Science, Section 6-6.3.   (Val has the master copy)
  • Marr, Vision: Sections 2.3-2.5
  • J. Feldman, Regularity-based perceptual grouping
    in Computational Intelligence, 13(4), 582-623, 1997. PDF
February 28 3D Vision, Geometric Image Properties
  • S&S: Chapter 12.
  • Forsyth and Ponce, "Computer Vision--A Modern Apprach"
    Chapter 5, "Geometric Image Features"

Homework 2 (due 3/15)

March 28 Camera Calibration, Pose Estimation
  • S&S: Sections 13.1-13.9
April 4 Multiple Views, Surface Reconstruction
  • S&S: Section 12.6
  • Forsyth and Ponce, "Computer Vision--A Modern Apprach"
    Chapter 12, "The Geometry of Multiple Views" (read all, but just skim 12.2-12.3 to become familiar with the math and notation)
  • Forsyth and Ponce, "Computer Vision--A Modern Apprach"
    Chapter 13, "Stereopsis" (again, read all, but skim 13.3 for the same reason)

Homework 3 (due 4/24)

April 11 Shape Cues, Light and Radiometry
  • S&S: Section 6.6, Section 12.3, Section 13.10
  • Forsyth and Ponce, "Computer Vision--A Modern Apprach"
    Chapter 2, "Radiometry--Measuring Light"
April 18 Motion Cues, Visual Tracking
  • S&S: Chapter 9
April 25 Image-Based Rendering, Virtual and Augmented Reality
  • S&S: Chapter 15
May 2 Project presentations

Readings

Required:

  • Shapiro, L. and Stockman G., Computer Vision, Prentice Hall, 2001.    (S&S)
  • Selected research papers and selections from other readings (below).

Other readings:
(many of these on reserve in the Math library)

  • Trucco, E. and Verri, A. Introductory Techniques for 3-D Computer Vision, Prentice Hall, 1998.
  • Horn, B., Robot Vision, MIT Press, 1986.
  • Forsyth, D. and Ponce, J., Computer Vision - A Modern Approach, on-line draft.
  • Marr, D. Vision, W.H. Freeman, 1983.
  • Regan, D. Human Perception of Objects: Early Visual Processing of Spatial Form Defined by Luminance, Color, Texture, Motion, and Binocular Disparity, Sinauer Assoc, 2000.

Links