Proposal
                  IEEE SIGNAL PROCESSING MAGAZINE
         Special Issue on Semantic Retrieval of Multimedia

There is no doubt that digital media has become an essential element
in all aspects of society.  Furthermore, the amount of digital media
that has been generated and stored, and which continues to do so at
an exponential rate, has already become unfathomable.  Much vigorous
research has focused on the problem of how to efficiently store,
search and retrieve digital content and in particular multimedia
content, such as images, video and audio.  The ease with which this
type of content is generated, coupled with its convenience and
flexibility, increases the emphasis on the need for its effective
access.  However, the majority of research up to now has taken a
relatively low-level approach to the multimedia indexing and
retrieval problem and by default has ignored higher level
interpretation and knowledge.  Consequently it has become blaringly
apparent that low-level approaches are proving to be extremely
limiting in making multimedia database systems accessible and useful
to end-users.  To this end, recent research has begun to focus on
bridging the semantic and conceptual gap which exists between man
and computer by integrating knowledge-based techniques, human
perception and understanding, high-level data interpretation, object
classification, psychology, language and semiotics.

The objective of the proposed special issue is to review and survey
new forward-thinking research and development in intelligent
content-based multimedia retrieval technologies. In addition issues
pertaining to limitations, sociological issues, future directions
and other related work is, of course, also welcome for publication.
Tutorial papers are to be solicited from the following
non-exhaustive list of topics:

Suggested Topics:
  * Video summarization (e.g., storybooks, skims, structure
    analysis, scene characterization)
  * Audio retrieval and classification  (e.g., music genre,
    song matching)
  * Human-computer interfaces and interaction to retrieval systems
    (e.g., intuitive interfaces) 
  * Semantic modeling for multimedia retrieval (e.g., language mapping
    to classified features and objects)
  * Object classification and retrieval in image databases
  * Sports and news video analysis and classification
  * Image database photo album and image storybook generation
  * Distributed multimedia indexing and retrieval 
  * Storage and accessibility issues (e.g., network infrastructure,
    hierarchical storage, distributed access)
  * Database visualization (e.g., statistical modeling)
  * Information mining for multimedia indexing and retrieval
  * Existing problems, roadblocks and future trends

Submission Procedure:

Prospective authors should submit white papers via the website http://www.cspl.umd.edu/spm/
			
  White paper due:          March 1, 2005
  Invitation notification:  March 15, 2005
  Manuscript due:           July 1, 2005
  Acceptance Notification:  October 15, 2005
  Final Manuscript due:     November 15, 2005
  Publication date:         March 2006

Guest Editors:

Anastasios N. Venetsanopoulos
Dept. Of Electrical Engineering
University of Toronto
Toronto, M5S 3G4, CANADA
anv@dsp.toronto.edu

Dimitrios Androutsos 
Dept. of Electrical Engineering
Ryerson University
Toronto, M5B 2K3 CANADA
dimitri@ee.ryerson.ca

Ling Guan
Dept. of Electrical Engineering
Ryerson University
Toronto, M5B 2K3, CANADA
lguan@ee.ryerson.ca