Language Resources for Content-based Image Retrieval
Held in conjunction with LREC'2008
There are two computer-aided methods for accessing images:
a.. indexing textual descriptions (done mostly by public retrieval engines)
a.. assigned manually on community sharing sites, e.g. Flickr;
b.. extracted from the text of the web page associated with the image , e.g. Google
c.. extracting textual information inside the image using OCR techniques
b.. indexing low-level visual features (colour, texture, etc.) of the images (called Content Based Image Retrieval - CBIR).
Combining these two methods is currently a hot topic since it involves developing hybrid approaches using automatic image tagging from text along with low level visual features from an image to produce text metadata. One of the main limitations within state-of-the-art CBIR resides in the mismatch (the "Semantic Gap" in image analysis) between low-level visual descriptors and the human-understood semantics of picture similarity. There are problems of vocabulary at all levels in the description of an image. There is a mismatch between the text found around an image, between the keywords that people assign to an image, and between the objects that can be recognized in an image. How can available language resources (terminology, thesauri, ontologies, etc.) help? How much must they be altered to reduce the semantic gap?
This half-day workshop will present up to 9 papers examining the intersection of language resources and image analysis and retrieval.
We welcome submissions addressing the following questions:
a.. What elements in a lexicon correspond to picturable objects?
b.. How can language resources be used to organize and improve user contributed image annotations?
c.. Which are the roles of language resources in image retrieval?
d.. How should language resources and image processing techniques be integrated?
e.. How can current language resources be adapted for content-based image retrieval in real-world applications, like the Web search?
f.. Can language resources and low-level visual descriptors be organized into a joint knowledge infrastructure?
g.. Do we need new semantic structures (ontologies, thesauri) for CBIR? Can we build them (semi)automatically?
h.. What visual characteristics of objects can be discovered by text processing (color, size, location, shape, texture)?
i.. What is the relation between the lexicon and image indexing and retrieval? What words should be associated with an image? Can this be automated?
j.. How to combine speech and textual resources related to image data?
k.. What are some application areas where this research can be evaluated?
We expect extended abstracts describing research addressing one of the above questions of about 1500 words to be submitted as WORD, PDF or ASCII documents per email to the following address: [log in to unmask]
A reception notification will be sent to the authors within 48 hours. The final papers should have no more than 6 pages.
Deadline for Abstracts
February 20, 2008
Notification of acceptance
March 10, 2008
March 31, 2008
May 26, 2008 (afternoon session)
Palais des Congrès Mansour Eddahbi, Marrakech (Morocco)
Co-located with LREC'2008 http://www.lrec-conf.org/lrec2008
Thierry Declerck, Deutsches Forschungszentrum für Künstliche Intelligenz, Germany
Adrian Popescu, Commissariat à l'Energie Atomique LIST, France
Allan Hanbury, Vienna University of Technology, Austria
Judith L. Klavans, University of Maryland, USA
Questions about the workshop can be addressed to: [log in to unmask] or [log in to unmask]
URL of the workshop: http://www.dfki.de/~declerck/ontoimage.html
The Workshop is endorsed among others by the European Network of Excellence K-Space (www.k-space.eu)
To unsubscribe, send an empty email to
mailto:[log in to unmask]
For further details of CHI lists see http://sigchi.org/listserv