Retrieval of Multimedia Content on the Web

Garant předmětu:

Lecturer:

Tutor:

Supervisor:

Department of Software Engineering

Synopsis:

The student obtains general knowledge regarding interfaces of portals providing multimedia content, the principles of similarity search, the methods of feature extraction from multimedia objects and indexing.

Requirements:

Basic knowledge of database systems, data structures, programming, internet technologies, calculus.

Syllabus of lectures:

1. Web platforms for retrieval and sharing of multimedia content.

2. Web interfaces, modalities, and the quality of retrieval.

3. Text-based and bag-of-words models.

4. Similarity search model - models and queries.

5. Similarity search model - popular similarity functions.

6. Image features - MPEG-7 descriptors.

7. Local image features.

8. Video retrieval techniques.

9. Feature extraction from audio, music and melody.

10. Retrieval of 2D shapes and 3D models.

11. Principles of metric indexing.

12. Metric access methods.

13. Unified model of retrieval in (non)metric spaces.

Syllabus of tutorials:

1. Project topic presentation.

2. Group consultations.

3. Group consultations.

4. Individual consultations.

5. Individual consultations.

6. Project presentation.

7. Project presentation.

Study Objective:

The module introduces the technologies of retrieval of multimedia content on the web. Since multimedia applications and databases acquire even larger space on the web, the module offers a perspective to the future developers of multimedia applications and databases - in and out of the web environment.

Study materials:

course slides +

1) Ricardo Baeza-Yates, Berthier Ribeiro-Neto. Modern Information Retrieval: The Concepts and Technology behind Search, 2011, Addison-Wesley Professional, ISBN-10: 0321416910

2) Phillipe Salembier, B. S. Manjunath, Philippe Salembier, Thomas Sikora. Introduction to MPEG 7: Multimedia Content Description Language, 2002, Wiley, ISBN-10: 0471486787

3) K. Selçuk Candan, Maria Luisa Sapino. Data Management for Multimedia Retrieval, 2010, Cambridge University Press, ISBN-10: 0521887399

4) Pavel Zezula, Giuseppe Amato, Vlastislav Dohnal, Michal Batko. Similarity Search: The Metric Space Approach, 2005, Springer, ISBN-10: 0387291466

Note:

Further information:

https://moodle-vyuka.cvut.cz/course/search.php?search=MIE-VMW

No time-table has been prepared for this course

The course is a part of the following study plans: