Bay Area OCR Interest Group

Bay Area OCR Interest Group

In the past few years the San Francisco Bay Area has become a hot spot for OCR and Document Analysis, with serious research and development efforts at a number of academic and corporate institutions. While much of what is going on is confidential, we believe that a certain amount of collaboration is also possible, even desirable, among the contributors in this area. The Bay Area OCR interest group (baocr) was formed to foster communication between researchers and developers of this technology.

Previous Meetings

  • November 6, 1995
    Thomas Breuel of IBM described his handwriting recognition work for the US Census Bureau and Larry Spitz of Daimler Benz talked about multilingual document recognition. SLIDES(gzipped PostScript 688k)

  • January 31, 1996
    We had a talk by Paul Gader on A Comparison of Fuzzy and Neural Methods for Street Number Location in Handwritten Addresses and a demonstration of Newton's new print recognizer by our host, Stuart Crawford.

  • April 18, 1996
    Dz-Mou Jung, Caere Corporation, Adaptive design of character classifiers in the field for higher accuracy
    Larry Yaeger, Apple Computer, Hand-print recognition in Newton SLIDES

  • September 30, 1996
    Jonathan J. Hull, Ricoh California Research Center, Content-Based Retrieval From Document Image Databases
    Tapas Kanungo, Caere Corporation, Automatic Generation of Character Groundtruth for Scanned Documents: A Closed-Loop Approach

  • October 24, 1996
    Thomas Bayer, Daimler Benz Research Center, Ulm, Germany Fresco: A Programming Language for Extracting Information from Structured Documents
    Andras Kornai, IBM Almaden Research Center, HMM Recognition of Personal Checks

  • March 13, 1997
    Gary Kopec, Xerox PARC Enlivening Legacy Documents in the UC Berkeley Digital Library Project
    Radovan Krtolica, Canon Research Learning character recognition by localized interpretation of character images

  • September 18, 1997
    Francine Chen and Dan Bloomberg, Xerox PARC Extraction of Indicative Summary Sentences from Imaged Documents
    Jianchang Mao, IBM Almaden Research Center Improving OCR Performance Using Character Degradation Models and Boosting Algorithm

  • February 11, 1998
    George Mills (reseach conducted at Apple Computer) Two Architectures for Segmenting Hand-printed Text
    Martin Boliek, Ricoh California Research Center Compression with Reversible Embedded Wavelets (CREW)

  • June 18, 1998
    Andreas Dengel, German Research Center for Artificial Intelligence (DFKI) OCR-Free Recognition and Labeling of Forms
    Dar Shyang Lee, Ricoh California Research Center A Neural Network Approach to Classifier Combination SLIDES

    Upcoming Meeting

  • March 1, 2000
    Henry S. Baird, Xerox PARC Document Image Analysis Research
    Yihong Xu, HP Labs Prototype Extraction and Adaptive OCR

    Place: Canon Research Center America, Inc. (CRA) 4009 Miranda Avenue, Palo Alto, CA 94304 Driving instructions

    For further information, or to have your name added to the baocr mailing list, please contact:
    Andras Kornai (
    Larry Spitz (

    Back to Home Page