Utrecht Summer School in Computer Science 07: Multimedia Retrieval
USCS Multimedia Retrieval
Course organizer: Remco Veltkamp
Description
After text retrieval, the next waves in web searching and multimedia
retrieval are the search for and delivery of images, music, video, and
3D scenes. Not only the perceptual and cognitive aspects, but also many
of the algorithmic and performance aspects are still badly understood.
One relevant issue is the design of dissimilarity measures (distance
functions) that have desired properties. Another aspect is the
development of algorithms that can compute or approximate these
distances efficiently. Indexing data structures and search algorithms
are necessary to make the search more efficient than sequential
browsing through large collections. Apart from provable properties of
individual algorithms, the experimental verification of the performance
of a complete retrieval system is important to analyse merits and
drawbacks of certain approaches, and to compare various techniques.
Target group
Graduate students (advanced master students and PhD students in the
early stage of their study) with a background in computer science.
Monday August 20
Topic: Issues in Multimedia Retrieval
Lecturer: Remco Veltkamp, Utrecht University
We will address perceptual issues, as well as formal properties of
features, similarity measures, and algorithms.
Tuesday August 21
Topic: Retrieval of Trademark Images
Lecturers: Helmut Alt, Ludmila Scharf, Sven Schulz, Free University Berlin
In trademark images, shape is a dominant feature. We will address shape similiarity measures and algorithms to compute them.
- Trademark retrieval and geometric matching [slides]
- Shape matching by random sampling [slides]
- Comparison of shapes in trademark images [slides]
Wednesday August 22
Topic: Retrieval of Music
Lecturers: Frans Wiering, Anja Volk, Jorg Garbers, Utrecht University
Two main groups of Music Information Retrieval (MIR) systems for
content-based searching can be distinguished, systems for searching
audio data and systems for searching notated music. This lecture will
primarily focus on searching notated music.
- Introduction to music information retrieval [slides]
- Musical content [slides]
- Music perception and cognition [slides]
- Modelling music similarity [slides]
- Rhythmic similarity based on inner metric analysis [slides]
- Music information retrieval systems [slides]
Thursday August 23
Topic: Retrieval of 3D Objects
Lecturers: Silvia Biasotti, Simone Marini, CNR-IMATI
Three-dimensional shape retrieval is fundamentally different from
two-dimensional shape retrieval. Most 2D methods do not generalize
directly to 3D. This is due to the different nature of the content:
descriptors used for 2D images are concerned with color, textures, and
properties that capture geometric details of the shapes segmented in
the image.
Introduction [slides]
What is a 3D shape [slides]
Similarity [slides]
Performance Evaluation [slides]
Shape matching methods [slides]
3D shape retrieval contest SHREC'07 [slides]
Conclusions [slides]
Friday August 24
Topic: Indexing
Lecturer: Reinier van Leuken, Utrecht University
Indexing data structures and search algorithms are necessary to make
the search more efficient than sequential browsing through large
collections.
Indexing [slides]
This summerschool is supported by the FP6 IST projects 511572-2 PROFI and 506766 AIM@SHAPE, and the Dutch ICES/KIS III
bsik project MultimediaN.
