Images and video play a crucial role in Visual Information Systems and Multimedia. There is an extraordinary large number of applications of such systems in entertainment, business, art, engineering, and science. Such applications often involve a client-server architecture, with large file and compute servers. Searching for images and video in large collections is becoming an important operation. Because of the size of such databases, efficiency is crucial.

We strongly believe that image and video retrieval need an integrated approach from fields such as image processing, shape processing, perception, data base indexing, visualization, querying, etc. On the other hand, most ongoing projects only deal with one or two of these aspects. A research emphasis is needed on incorporating multiple models for shape, color, texture, geometry, and syntax, so that the user does not have to specify low-level model parameters and combinations. This should lead to strategies of efficient indexing of visual information, in addition to techniques for combining visual with more traditional database information. Realistic evaluation criteria are needed, including test databases of realistic size in domains of interest, measures of similarity that allow variations in perceptual, semantic, and other criteria, and measures of accuracy and efficiency in assisting the user.

The purpose of this first Dagstuhl Seminar "Content-Based Image and Video Retrieval" was to bring together people from the various fields in order to promote information exchange and interaction among researchers who are interested in various aspects of accessing the content of image and video data, including topics such as:

• Indexing schemes
• Matching algorithms
• Visual data modeling
• Retrieval system architectures
• Image and video databases
• Feature recognition
• Video segmentation
• Picture representation
• Query processing
• Perception issues
• Video and image compression
• Visualizing pictorial information
• Searching the web
• Delivery of visual information
• Benchmarking
• Application areas of image and video retrieval

For this seminar, we have invited internationally known as well as young researchers from various disciplines with a common interest in content-based image and video retrieval. We have been together with a group of 28 researchers for a week, away from the rest of the world, and certainly good interaction and exchange of ideas took place during the sessions as well as in the very "gemütliche" wine cellar, enjoying the cheese platter.

There was a total of 24 presentations, two demonstration sessions, and two discussion sessions. The abstracts of the presentations are presented here, in the order in which they have been written by hand in the book of abstracts. One discussion session was about the challenges and problems to be solved, the other session was about the particular problem of how to assess the quality of retrieval systems and algorithms for subtasks, see the report at the end of this booklet.

• Hans Burkhardt, Freiburg Universität
• Hans-Peter Kriegel, Universität München
• Remco Veltkamp, Utrecht University