Missing feature preprint available

[Subutai Ahmad]

The following paper is available for anonymous ftp on Neuroprose,
archive.cis.ohio-state.edu (128.146.8.52), in directory
pub/neuroprose, as file "ahmad.missing.ps.Z":


Some Solutions to the Missing Feature Problem in Vision

Subutai Ahmad and Volker Tresp
Siemens Central Research and Development

Abstract

In visual processing the ability to deal with missing and noisy
information is crucial.  Occlusions and unreliable feature detectors
often lead to situations where little or no direct information about
features is available.  However the available information is usually
sufficient to highly constrain the outputs.  We discuss Bayesian
techniques for extracting class probabilities given partial data. The
optimal solution involves integrating over the missing dimensions
weighted by the local probability densities. We show how to obtain
closed-form approximations to the Bayesian solution using Gaussian
basis function networks. The framework extends naturally to the case
of noisy features. Simulations on a complex task (3D hand gesture
recognition) validate the theory. When both integration and weighting
by input densities are used, performance decreases gracefully with the
number of missing or noisy features. Performance is substantially
degraded if either step is omitted.

To appear in:

S. J. Hanson, J. D. Cowan, and C. L. Giles (Eds.), Advances in Neural
Information Processing Systems 5. San Mateo CA: Morgan Kaufmann.


----
Subutai Ahmad
Siemens AG, ZFE ST SN61,                        Phone: +49 89 636-3532
Otto-Hahn-Ring 6,                                 FAX: +49 89 636-2393
W-8000 Munich 83, Germany           	  E-mail: ahmad at zfe.siemens.de