Gregory J. Clary

Learn More
The curling of a graphitic sheet to form carbon nanotubes produces a class of materials that seem to have extraordinary electrical and mechanical properties. In particular, the high elastic modulus of the graphite sheets means that the nanotubes might be stiffer and stronger than any other known material, with beneficial consequences for their application(More)
USING STATISTICAL METHODS Krishna S. Nathan, Homayoon S. M. Beigi, Jayashree Subrahmonia, Gregory J. Clary and Hiroshi Maruyama Handwriting Recognition Group IBM Research, T.J. Watson Center, Yorktown Heights, NY 10598 ABSTRACT We address the problem of automatic recognition of unconstrained handwritten text. Statistical methods, such as hidden Markov(More)
This paper will emphasize the necessity of having alternative man-machine interfaces to the already established keyboard and mouse, speciically for people residing in developing countries. An on-line system is presented for recognizing handwritten Farsi (Persian) and Arabic digits. This recognition scheme is based on statistical techniques which will be(More)
PURPOSE To determine the speed, accuracy, ease of use, and user satisfaction of various electronic data entry platforms for use in the collection of mammography clinical trials data. METHOD AND MATERIALS Four electronic data entry platforms were tested: standalone personal digital assistant (PDA), Tablet PC, digitizer Tablet/PDA Hybrid (DTP Hybrid), and(More)
I define a novel class of medial primitives called anchor primitives to provide a stable framework for feature definition for statistical classification of image sequences of closed objects in motion. Attributes of anchor primitives evolving over time are used as inputs into statistical classifiers to classify object motion. An anchor primitive model(More)
  • 1