Summary: Shape Quantization and Recognition
with Randomized Trees
Yali Amit and Donald Geman y
Department of Statistics, University of Chicago, Chicago, IL, 60637; Email: firstname.lastname@example.org.
Supported in part by the ARO under grant DAAL-03-92-G-0322.
yDepartment of Mathematics and Statistics, University of Massachusetts, Amherst, MA 01003;
Email:email@example.com. Supported in part by the NSF under grant DMS-9217655, ONR under
contract N00014-91-J-1021, and ARPA contract MDA972-93-1-0012.
We explore a new approach to shape recognition based on a virtually in nite family
of binary features (\queries") of the image data, designed to accommodate prior in-
formation about shape invariance and regularity. Each query corresponds to a spatial
arrangement of several local topographic codes (\tags") which are in themselves too
primitive and common to be informative about shape. All the discriminating power
derives from relative angles and distances among the tags. The important attributes of
the queries are (i) a natural partial ordering corresponding to increasing structure and
complexity; (ii) semi-invariance, meaning that most shapes of a given class will answer
the same way to two queries which are successive in the ordering; and (iii) stability,