Correspondence Problems in Computer Vision
Computer Science Teaching Award (summer term 2008)
Lecturer:
Dr. Andrés Bruhn
Office hours: Friday, 14:15 - 15:15.
Summer Term 2009
Lectures (2h) with programming/theoretical exercises (2h)
(6 credit points)
Lectures: Wednesday 14-16 c.t., Building E1.3, Lecture Hall 1
First lecture: Wednesday, April 29, 2009.
Tutorials: 2 hours each week, 2 groups: Thursday 16-18 and 18-20 c.t.
Turorials: Building E1.3, SR 015 (theory)
Turorials: Building E2.4 (math building), Cip-Pool U009 (programming)
First tutorial: Thursday, May 7, 2009.
Prerequisites –
Synopsis –
Planned Contents –
Assignments –
References
Example - Stereo Reconstruction
Example - Motion Estimation
Requires undergraduate knowledge in mathematics (e.g. ''Mathematik
für Informatiker I-III'') . Knowledge in image processing or differential
equations is useful. The lectures will be given
in English.
Correspondence problems are a central topic in computer vision. Thereby,
one is interested in identifying and matching corresponding
features in different images/views of the same scene. Typical corresondence
problems are the estimation of motion information from consecutive frames
of an image sequence (optic flow), the reconstruction of a 3-D scene from
a stereo image pair and the registration of medical image data from
different image acquisition devices (e.g. CT and MRT). Central
part of this lecture is the discussion of the most important correspondence
problems as well as the presentation of suitable algorithms for solving them.
This class is particularly useful for those students who wish to
to pursue a diploma or master thesis in our group in the field of
computer vision.
Date | Topic
|
29/4 |
Introduction, Overview |
6/5 |
Block Matching, Correlation Techniques, Interest Points, Feature-Based Methods |
13/5 |
Optic Flow I: Local Differential Methods, Parameterisation Models |
20/5 |
Optic Flow II: Global Differential Methods, Horn and Schunck |
27/5 |
Optic Flow III: Advanced Constancy Assumptions, Large Motion |
3/6 |
Optic Flow IV: Advanced Data and Smoothness Terms |
10/6 |
Optic Flow V: High Accuracy Methods, Advanced Numerics |
17/6 |
Stereo Matching I: Projective and Epipolar Geometry |
24/6 |
Stereo Matching II: Estimation of the Fundamental Matrix |
1/7 |
Stereo Matching III: Correlation and Variational Methods, Graph Cuts |
8/7 |
Medical Image Registration: Mutual Information, Elastic and Curvature Based Registration, Landmarks |
15/7 |
Particle Image Velocimetry: Div-Curl-Regularisation, Incompressible Navier Stokes Prior |
Programming excercises and theoretical assignments are offered as part of the tutorials.
The regular attendence of these excercises is requirement for admission to
the exam.
These correct solutions are partially based on the solutions in LaTex provided by Jan-Hendrik Dithmar
(thanks!). If you encounter any problems, please notify the lecturer.
The first written exam took place on Wednesday, July 22 from 2 to 4 PM
in Building E13, Lecture Hall 001.
The second written exam took place on Thursday, September 24 from 2 to 4 PM
in Building E13, Lecture Hall 002.
The best one of both grades counts.
The grades for the second written exam are available!
You can inspect your exam sheets on Friday, October 9, 14:00-15:00,
building E1.1, room 3.06 (3rd floor).
The following thresholds were applied to determine the grades:
- 1.0 : 30 - 28 points
- 1.3 : 27 - 27
- 1.7 : 26 - 25
- 2.0 : 24 - 24
- 2.3 : 23 - 22
- 2.7 : 21 - 21
- 3.0 : 20 - 19
- 3.3 : 18 - 18
- 3.7 : 17 - 17
- 4.0 : 16 - 15
- 5.0 : 14 - 0
The detailed distribution of points was:
- 27 points : 1
- 24 points : 3
- 23 points : 2
- 22 points : 1
- 21 points : 2
- 20 points : 1
- 16 points : 1
- 15 points : 1
- 14 points : 1
- 13 points : 1
- 12 points : 1
The results can be queried via our online query form.
The cerfificates (Scheine) are issued by the office of the Mathematics Department. They can be obtained from
Mrs. Voss, Building E2.4, Room 111 (math building, ground floor, 8.15-11.30 AM).
There is no specific book that covers the complete content of this class.
Many lectures will be based on articls from journals
and conferences. However, the following three books cover most
topics:
- Optic Flow
A. Bruhn: Variational Optic Flow Computation: Accurate Modeling and Efficient Numerics.
Ph.D. Thesis, 2006. Available from /bruhn/PhDThesis.pdf
- Stereo Reconstruction
O. Faugeras and Q.-T. Luong:
The Geometry of Multiple Images.
MIT Press, 2001.
- Medical Image Registration
J. Modersitzki:
Numerical Methods for Image Registration.
Oxford Press, 2003.
|