IRG 3 - Teaching
Data Processing Tips and Tricks, WS 2005/2006
Specialized Seminar - Computer Graphics / Computer Vision
Description:
In completely different areas of computer science, engineering, and
natural sciences, one frequently stumbles across the same algorithms
and techniques. Chances are that you haven't yet heard much about most
of them. Knowing these methods, however, might come in very handy once
you have started working on your thesis in computer vision, graphics,
machine learning, data mining, or any other field having to do with
some form of measured data.
In this seminar, you will find out everything you always wanted to
know about
- Bayesian statistics
- EM estimation
- wavelets
- Levenberg-Marquardt optimization
- Lloyd clustering
- Karhunen-Loeve transform
- singular value decomposition
- multi-dimensional scaling
- algebraic reconstruction technique
- support vector machines
- ...
but were afraid to ask. Main emphasis will be put on how to successfully
apply these methods in practice. The goal is to gain an intuitive
understanding of how each technique works, and what its capabilities and
limitations are.
Time and Place:
- Friday, 11 - 13, MPII, Room 019 (Visualization Room)
- first lecture: October 21, 2005
Instructors:
Requirements:
To successfully participate in this seminar, you have to prepare a web
site describing (a) what your technique is good for, (b) how it works in
theory, and (c) how it is applied in practice. Give a lot of examples from
many different fields. Show lots of insightful figures. Include example
code (preferably in C/C++) that later on everyone can download, run, and
understand. Use your web page and code to describe your technique during
you presentation in the seminar.
Prerequisites:
Remarks:
- either English or German, as preferred
Course Notes:
Seminar Wiki - Homepage preparation
- Seminar
Wiki - Every student has already a page assigned to her /
his topic
- Review Form - An XML Form intended for
the reviews
Topic Assignment:
- Algebraic reconstruction technique (Lukas)
- Martin Emrich
- Dorotea Dudas
- Christine Walter
- Wavelets (Lukas)
- Sabrina Linn
- Sascha El-Abed
- Rainer Jochem
- Matthias Berg
- Graph Cut Optimization (Timo)
- Art Tevs
- Iliyan Georgiev
- Irina Brudaru
- Bayesian statistics (Timo)
- Ciprian Raileanu
- Rostislav Rusev
- Stefana Nenova
- Support Vector Machines (Timo)
- Daniel Fischer
- Konstantin Halachev
- Harriet Bach
- Level-Set Methods (Ivo)
- Nazar Khan
- Marie-Christin Dieter Anthony
- Dmitrij Tsesarskij
- RANSAC (Ivo)
- Manuel Gorius
- Orhan Sönmez
- Evren Ermis
- Levenberg-Marquardt optimization (Ivo)
- Steffen Heil
- Roman Brauchle
- Jasmina Bogojeska
- EM Algorithm (Volker)
- Yana Mileva
- Patrice Rousseau
- Johannes Hoen
- Singular Value Decomposition (Andrei)
- Manuel Schilz
- Michael Arnold
- Andreas Gross
- Sebastian Wilhelmi
- Lloyd clustering (Andrei)
- Gaurav Pandey
- Markus Hoffmann
- Olivier Engelkes
- Carlos Figueredo
- Dennis Stachowicz
- Karhunen-Loeve transform -- Principal Component Analysis
(Christian)
- Tomasz Wegrzanowski
- Kristina Scherbaum
- Peter Schaefer
References:
