Computing That Serves

A New Paradigm for Experimental Pattern Recognition Research


Thursday, September 16, 2010 - 11:00am


Daniel Lopresti
Professor and Chair
Department of Computer Science and Engineering
Lehigh University (Bethlehem, PA)


Eric Ringger

Despite tremendous advances in computer hardware and software, the manner in which we conduct experimental pattern recognition research has remained essentially unchanged for 50 years.  In this talk, we discuss serious issues with the old model which place constraints on the field's ability to make progress.  We then outline a vision of the future where community-maintained resources and Web 2.0 collective intelligence make possible fundamental changes in the experimental science.  This new paradigm allows for multiple interpretations in place of a single "ground-truth," eliminates explicit and implicit bias in testing, rewards solutions that are general and robust, and promotes reproducibility of experimental results.  Ultimately, it will make possible the asking of big-picture questions that are now well out of reach.

In support of a prototype implementation which we expect to field within the next year, we describe current work at Lehigh and elsewhere that is helping to lead us in this direction.  We illustrate the discussion with examples drawn from document analysis and noisy text analytics.  Finally, we conclude with suggestions for those wishing to become involved.


Daniel Lopresti received his bachelor's degree from Dartmouth in 1982 and his Ph.D. in computer science from Princeton in 1987.  After completing his doctorate, he joined the Department of Computer Science at Brown and taught courses ranging from VLSI design to computational aspects of molecular biology and conducted research in parallel computing and VLSI CAD.  He went on to help found the Matsushita Information Technology Laboratory in Princeton, and later also served on the research staff at Bell Labs where his work turned to document analysis, handwriting recognition, and biometric security.

In 2003, Dr. Lopresti joined the Department of Computer Science and Engineering at Lehigh where he leads a research group examining fundamental algorithmic and systems-related questions in pattern recognition, bioinformatics, and computer security.  He has authored or co-authored over 100 publications in journals and refereed conference proceedings and holds 21 U.S. patents.  He is a founding co-chair of the Noisy Text Analytics workshop series, has twice co-chaired the IAPR Workshop on Document Analysis Systems, and is a co-chair for the 2011 and the 2013 International Conference on Document Analysis and Recognition.  Dr. Lopresti is co-director of the Lehigh Pattern Recognition Research Lab, and on July 1, 2009, he became Chair of the Department of Computer Science and Engineering.