Important Dates:

November 2-4, 2005

All restricted sites are intended for SDIUT'05 Participants ONLY do not repost.

SDIUT05 Flyer: SDIUT05.pdf

SDIUT05 Agenda: SDIUT05-Agenda2.pdf

SDIUT05 Participants:
sdiut05a.pdf Restricted

SDIUT05 Proceedings: Symposium.pdf Restricted

SDIUT05 Proceedings: Presentations

Additional Sponsorship:

Contact Information:

C/O Denise Best
University of Maryland
College Park, MD 20742

Fax: (301) 314-2644
Phone: (301) 405-6444

To be added to the SDIUT mailing list, send email to the address above

Previous SDIUT:

SDIUT03, Greenbelt, MD
, Columbia, MD
SDIUT99, Annapolis, MD














Did you forget something at SDIUT05 in the conference room?
If so please contact Denise Best at
to make arrangements for its safe return.


If you would like your presentation made available to attendees of the Symposium, please submit them here SDIUT05 Presentation Submissions.




Who should Particpate?... Click Here

Call for Participation:


Leading researchers from industry and academia, as well as all government agencies involved in document understanding, are being invited to a three-day Symposium on Document Image Understanding Technology (SDIUT05) to be held at the The Marriott Inn and Conference Center, University Maryland University College, Adelphi, Maryland, on November 2-4, 2005. Funding agencies will describe present and future document understanding programs and principal investigators will report on current efforts in the field.
This Symposium is sponsored in part by federal agencies funding document research and will serve as an overview of state-of-the-art projects in document image understanding.

Topics include, but are not limited to:

* Multi-lingual OCR
* Language Identification
* Document Layout Analysis
* Logical Structure Recognition
* Document Classification
* Standards for Document Understanding
* Document Databases
* Recognition of Degraded Documents
* Tools for Document Analysis
* Commercial Systems

Special emphasis will be placed on the effects of degradation and document analysis processes on:

*Forms Processing and Logo Detection/Recognition
*Machine Translation
*Information Extraction
*Image and Video Text Recognition
*Information Retrieval
*Cross Language Retrieval
*Document Image Management
*Electronic and WWW Document Analysis

We are very interested in expanding the topics presented to new areas. To do this, we are inviting all attendees to consider participating as speakers during an abstract session, or to consider bringing a poster or demo. This is an ideal way to introduce your work or area of interest to the community.

Who should participate in SDIUT?

Government Agencies and Representatives: Representatives are invited to participate by giving a description of a document image analysis need, program or project, or by attending and interacting with other researchers and developers. In the past we have had representatives brief the community from many agencies and hope to include additional representatives to discuss document analysis needs that have evolved over the past several years.

University Researchers: Cutting edge research is a key component to solving many of the government's unique problems. This symposium gives researchers the opportunity to communicate developing ideas that show promise for integration into large volume and often highly degraded document streams. Technical presentations as well as posters and demonstrations are welcome.

Companies: SDIUT provides a unique opportunity to interact with government researchers and demonstrate capabilities in research and development or engineered solutions. We will be continuing our focus on interaction with a demonstration and poster session (and reception) on Thursday afternoon, November 3rd. Leading OCR and document analysis companies will demonstrate their products. There is no additional charge for participation.