TSBB13 Computer Vision Systems

The course gives 6 ECTS credit points, which corresponds to approximately 160h of work per student.  This time is divided into the following activities

• Lectures: 4 × 2h = 8h
• Seminars: 2 x 2h = 4h 
• Own work in projects: 148h

The course is given during the VT2 period.

Messages 

• 2010-04-22: The lecture scheduled for April 26 has been moved to May 12, between 13-15.

People

• Klas Nordberg, lectures, examiner
• Per-Erik Forssén, lectures
• Fredrik Larsson, project guide
• Johan Hedborg, project guide

We all have our offices in the B-building, corridor A between entrances 27 and 29, ground floor.

Documents

• LiTH Study Guide for TSBB13, 2010
• Course schedule

Literature 

The course is based on the book

• Sonka, Hlavac, Boyle, Image Processing, Analysis, and Machine Vision, 3rd edition, Thomson (2008).

There is a student (paperback) edition of this book (labeled ISE) from Nelson Engineering. Further material will be provided during the course.

Registration

In order to receive email about the course and have results input to Ladok it is required that you are registered for the course.  If you intend to take the course but is not registered, make sure to register asap, using the Student portal.  If you take the course but are not registered to any program at Linköping University, please contact the course examiner in order to make sure that you receive email about the course.

Examination

Each student participates in a smaller group (ca 3 students per group) and the examination is primarily based on the results produced by the group in relation to the expected outcomes.  Each student is expected to actively participate in the groups and to document contribution to the results of the group.  Written and oral presentations of the group's results are expected to be shared among the group participants. Grades will be given according to the criteria in the study guide. If a group did not pass the written and oral presentation at the respective seminar, a new date for the project presentation will be decided by the examiner. Supervision after the second seminar cannot be guaranteed.

Report

All groups should produce one report per project that documents the work.  The report should be brief, max 6 pages, but must contain description of which problem has been addressed, how you tried to solve it, and what are the results and conclusions.  It should contain sufficient information to reproduce your work, i.e., give values to parameters, etc.  Provide relevant references, e.g., to methods that you are using.  The report is due on the seminar when the project is presented.  It is expected that all students in a project group contribute to all reports produced by the group.

Higher grades

To get higher grades than "passed" (grade 3), the following guidelines applies:

• Higher grades are given based only on individual work.
• Grade 4 is given to students that can make a quantitative comparison between the methods they have implemented in their own project with different methods implemented by other groups on the same data.  Alternatively, a quantitative comparison of own methods on a broader set of data than the standard set supplied for the basic examination, e.g., produced by the student.  Quantitative comparison mean here to describe what the relevant differences or similarities are, but not to analyse what causes the differences/similarities.
• Grade 5 is given to students that, beyond the grade 4 requirements, can produce a relevant analysis of the reasons for the result of the quantitative analysis.  This means here to describe the reasons for the differences/similarities.

Both projects should be covered to pass with grade 4 or 5.  If ony one project is covered, grade 4 is given if both a quantitative comparison and a relevant analysis is made.

The results of these investigations should be documented in a brief report.

Grades will be reported at latest 10 days after the last seminar.  Additional work for higher grades will be reported at the stipulated examination periods.

Subversion

All project groups have access to a subversion based repository for management of program code and project documentation and is expected to use it for this purpose.

A subversion client for Windows based platforms can be downloaded from here.

Schedule for the lectures

The two last seminars are obligatory for all course participants in order to pass the examination.

Participants

The project groups are formed during the first lecture and it is therefore important to attend this lecture.  If you plan to participate in the course but cannot attent the first lecture, send an email to the examiner, otherwise participation in the course cannot be guaranteed.

Here is a list of the groups and participants

Projects

There is a page with useful hints and pitfalls for the projects

Additional material

These articles should be accessible from LiU IP-addresses

• Triggs, McLauchlan, Hartley & Fitzgibbon, Bundle Adjustment - A Modern Synthesis, Proceedings of Vision Algorithm: Theory and Practice, 2000, doi:10.1007/3-540-44480-7_21
•  Hartley & Sturm, Triangulation, Proceedings of Computer Analysis of Images and Patterns, 1995, doi:10.1007/3-540-60268-2_296

Calendar

• No item yet