B. Programing prerequists:

- General programing skills. JAVA itself is not essential. The essential skills are:

-- Knowledge of simple datastructures.

-- Ability to construct programs.

-- Ability to assess on the efficiency and the correctness of a program.

-- A structured approach to testing of small programs.

- Matlab programming skills.

Matlab on a level which can be achieved at the course \"Signal processing\". That is the ability to design and implement small Matlab programs and the ability to understand, explain and make minor modifications to large Matlab programs.

C. Signal processing prerequists:

The student can answer the following to a great detail including describing the underlying math: What is a signal, a measurement, noise, the difference between an analog world and the digital world, one dimensional convolution, correlation, filtering, low/high/bandpass filtering.

Introductory programming (GP) or equivalent which cannot be followed simultaneously.

You should be able to write larger programs in either matlab or C++.

The Image analysis course is not a prerequsite for this course but is strongly adviced to take it prior or at the same time. 

Focus is on the elements for making computer vision systems which are able to exploit the relations between objects in the real world and the images taken by one or more cameras or in movies.

The course will be slightly bias towards using computer vision for human computer interaction. That is, using cameras connected to a computer, how can we make a computer program which is able to localize, recognize and follow human gestures so that the user can interact with the computer.

After completion of this course, you will be able to:

Understand, choose between, and apply various computer vision algorithms.

Understand the relations between objects in the 3D world and those obtained from cameras.

Understand the principles on how to make 3D models (reconstruction) from images.

Write programs which are able to follow objects in pre-recorded movies or live images obtained from cameras in either Matlab or C++.

Understand principles for making computer vision systems that aim towards enabling humans to interact with a computer through cameras. 

A total of 3-4 larger mandatory assignments will given for the exercises. Examples will be drawn for HCI, industrial applications and medical imaging.

The exercises focus on hands-on programming experience with computer vision problems, through the larger assignments and aims towards completing a set of common computer vision problems.

The topics you will be introduced to on this course are:

Object tracking

Object Recognition

Camera calibration

Stereo vision

Multiple view geometry (Geometrical / projective relations and transformations between images).

Computer vision for human computer interaction

How to develop computer vision applications 

Notice: As described in the course title, this course is planned to be an evening course with lectures from 5 pm to 7 pm and exercises from 7 pm to 9 pm. But: lectures might run in the daytime (9 am to 12 am and exercises from 1 pm to 4 pm)if all students wish.

NB! In the introductory week, meaning from 27 August to 2 September 2004 the course will run as usual, meaning: Lectures from 5 pm to 7 pm and exercises form 7 pm to 9 pm.
 

The 3-4 mandatory assignments constitute the foundation for the final report to be handed in 19 November 2004 12:00 at the Examination Office.