EURASIP SIGNAL PROCESSING Journal
(http://www.elsevier.nl/locate/sigpro)

Special Issue on: Content-based Visual Information Retrieval

Content-based retrieval of visual information is an emerging technology that extends
traditional information retrieval to data repositories containing visual information, such
as images or video. It combines aspects of signal processing, computer vision, machine
learning and information retrieval and has applications in multiple settings including
consumer electronics, web search engines, surveillance, visualization and data management.

While considerable progress has been achieved over the past few years, a significant
number of challenging questions are still wide open. These include issues in 1)
representation, e.g. the quest for perceptually justified similarity, 2) learning, e.g.
automated extraction of semantic descriptors, 3) theoretical characterization of retrieval
algorithms, e.g. convergence guarantees for relevance feedback algorithms, 4)
high-dimensional indexing, 5) interface design, e.g. bridging the semantic gap between
query by example and natural language queries, 6) information fusion, e.g. optimal
techniques for integrating information from diverse content modalities and 7) event
detection and recognition in video streams. The purpose of this special issue is to survey
the latest developments in both the theory and practice of visual information retrieval.
For this, original high-quality contributions are solicited in, but not restricted to, the
following topics:
                -       Representation: features and models
                -       Visual similarity functions
                -       Relevance feedback algorithms
                -       Extraction of semantic content descriptors
                -       Aspects of machine learning in content-based retrieval
                -       High-dimensional indexing
                -       Integration of diverse multimedia sources
                -       Motion analysis and characterization
                -       Mosaics and video layering
                -       Spatial and temporal video segmentation
                -       Event detection and classification
                -       Video summarization
                -       Image and video retrieval engines
                -       User-interface issues
                -       Unified representations for compression and retrieval
                -       Perceptual studies and perceptually motivated systems
                -       Applications: clinical databases, arts, surveillance, etc.

Prospective authors should submit 6 hard-copies and an electronic copy of their manuscript
to the address below, according to the following timetable.

  Manuscript due:           June 28, 2002
  Acceptance notification:  October 31, 2002
  Final Manuscript due:     November 15, 2002

To qualify for publication, papers must be previously unpublished and not be under
consideration for publication elsewhere. Upon acceptance of an article, the author(s) will
be asked to transfer copyright of the article to the publisher. This transfer will ensure
the widest possible dissemination of information.
Accepted language is English. The text of the paper should be preceded by an abstract of
no more than 200 words. Abstracts should contain the substance of the methods and results
of the paper. Page proofs will be sent to the principal author. All pages should be
numbered. The first page should include the article title and the author's name and
affiliation, as well as a name and mailing address to be used for correspondence and
transmission of proofs. Manuscripts should not exceed 40 double space pages, including all
figures and references.

GUEST EDITOR
Dr. Nuno Vasconcelos, 
Cambridge Research Laboratory, Compaq Computer Corporation, 
One Cambridge Center, Cambridge MA 02142; 
nuno.vasconcelos@compaq.com