Multimedia Retrieval
On this page you find information about the course "Multimedia Retrieval", which is given by Andreas Nürnberger in summer term 2017. This page will be updated regularly throughout the semester.
General Information
Multimedia Retrieval (MIR) considers information search in non or less structured multimedia data sets. Examples for applications are web search engines, digital libraries and multimedia archives like music, picture or video data bases. As part of this course the basic MIR concepts will be introduced and discussed based on diverse application examples. Since the content based search is important, also different methods for representing and indexing content of texts and multimedia objects will be presented.
Requirements for Participation in the Final Exam
There will be exercise sheets for the exercises. You need to "vote" for exercise tasks. By "voting" for a task the student shows that she / he is able to present this task in front of the class. (Possible solutions will be discussed. They do not need to be correct. However, we expect that the "voter" is well prepared and is able to present at least an initial solution in front of the class.) The tasks will be split up into 2 groups: programming tasks and other tasks. The programming tasks will be marked as such.
The requirements for participation in the final exam (oral or written) are:
- at least 66% of all programming points,
- at least 66% of votes for all other exercise tasks and
- at least one time presentation of a solution in front of the class.
The requirements for the "Schein" (without a mark) are pass of a short colloquium of approximatly 10 minutes or pass of the written exam at the end of the semester. (If an oral or written exam is offered then it will be announced on this page.)
Dates and Rooms
Time | Start | Room | |
Lecture | Mo, 15:15-16:45 o'clock | 03.04.2017 | G22A-110 |
Exercise |
Tu, 09:15-10:45 o'clock Mo, 11:15-12:45 o'clock |
11.04.2017 10.04.2017 |
G22A-210 G22A-128 |
Further information about the lecture and exercise can be found in the LSF portal.
Teaching Staff
If you have any questions about the lecture or the exercises, please contact us via e-mail:
Lecture:
Exercise:
Materials
Lecture Slides
- 00. General Course Info
- 01. Introduction
- 02. Vector Space Model
- 03. Evaluation
- 04. Structuring
- 05. MPEG7 Overview
- 06. MPEG7 Visual Descriptors
- 07. Video-Retrieval
- 08. Audio-Retrieval
- 09. MPEG7 Audio Descriptors
- 10. Human-Computer-Interaction
- 11. Algorithms and Data Structures
Exercise Sheets
- Exercise 01 (until 10.04.2017 respectively 11.04.2017)
- Exercise 02 (until 25.04.2017) also see "Additional Material" below
- Exercise 02 also contains the assignment for the proramming task P01 (Deadline: 16.05.2017)
- Programming task evaluation
- Exercise 03 (until 02.05.2017)
- Exercise 04 (until 09.05.2017)
- Exercise 05 (until 16.05.2017)
- Exercise 06 (until 23.05.2017)
- Exercise 07 (until 30.05.2017)
- Exercise 08 (until 13.06.2017)
- Exercise 09 (until 20.06.2017)
- Exercise 10 (until 27.06.2017)
- Exercise 11 (until 04.07.2017)
- Assignment for the programming task P02 (Deadline: 13.06.2017)
- also see "Additional Material" below
- Programming task evaluation
Additional Material
- Inverted Index + Lucene Quick Tutorial
- mpeg7bindings.zip
- mpeg7bindings_usage.pdf
- Pearson Algorithm.jpg
- gioconda.bmp
- AudioExample.zip
- ...
Literature
- Ähnlichkeitssuche in Multimedia-Datenbanken
Ingo Schmitt
Oldenbourg Wissenschaftsverlag GmbH, München, 2005. - Modern Information Retrieval
Ricardo Baeza-Yates and Berthier Ribiero-Neto
Addison Wesley, 1999. - Foundations of Statistical Natural Language Processing
Chris Manning and Hinrich Schütze
MIT Press, Cambridge, MA, 1999. - Information Retrieval: Data Structures and Algorithms
William B. Frakes and Ricardo Baeza-Yates
Prentice-Hall, 1992. - Soft Computing in Information Retrieval
Fabio Crestani and Gabriella Pasi
Physica Verlag, 2000.
Links to other Web Pages
- Information Retrieval book by C. J. van Rijsbergen
- Information Retrieval journal, Kluwer
- ACM Special Interest Group on Information Retrieval