Since 2010, we have introduced the Python programming languge and the corresponding libraries for image processing and computer vision. Python has become more and more popular as a replacement for Matlab in the Computer Vision community. What's more, it is free!

Another incentive for learning Python:
LucasFilm Singapore wants CS students who also know Python, ML and unix/linux scripting.
 
  This module investigates methods for computers to understand and interpret the contents of images and videos. It aims at teaching the modern concepts underlying various computer vision techniques and the application of current techniques to problem solving. The topics covered in this module include image registration and mosaicking, feature detection and matching, tracking, 3D vision, camera calibration and 3D reconstruction, pattern recognition, etc. Laboratory exercises and projects are included for the students to learn to use computer vision and image processing software packages, and to gain experience in solving interesting practical problems.
 
  > Understand the basic concepts underlying various computer vision techniques.
> Able to analyse computer vision problems and to apply computer vision techniques to solve the problems.
> Able to use computer vision software tools to implement the solutions to the problems.
> Able to read, understand, and implement other related algorithms outside the scope of the lecture materials.
> Able to work in a team to take on an interesting computer vision project.
 
  MA1101R Linear Algebra or MA1506, MA1505/1505C Calculus, and ST2131 Statistics.

Mathematics is an essential tool for the study of computer vision and pattern recognition. You should have enough mathematics background, particularly undergraduate linear algebra, calculus, and a bit of probability. For a sample of the mathematics used, check sample-math.pdf.
 
 
lab excercises
15%
written assignments
10%
project
35%
final exam (open book)
40%
total
100%

Your current CA grades are available here.
 
 

There are several reasonably good textbooks on computer vision. This module uses mainly the materials in the following books:
>> Szeliski, Computer Vision: Algorithms and Applications, Springer, 2010 (soft copy available in author's website).
>> Bradski and Kaehler, Learning OpenCV: Computer Vision with the OpenCV Library, O'Reilly, 2008.

Other nice references include:
>> Forsyth & Ponce, Computer Vision: A Modern Approach, Prentice Hall, 2003.
>> Shapiro & Stockman, Computer Vision, Prentice-Hall, 2001.
>> Jain, Kasturi, and Schunck, Machine Vision, McGraw Hill, 1995.
 

  A/Prof. Leow Wee Kheng
Email: leowwk@comp.nus.edu.sg
Web: www.comp.nus.edu.sg/~leowwk

Last update: 12 July 2012