Vector quantization methods are confronted with a model selection problem, namely the number of prototypical feature representatives to model each class. In this paper we present a...
Alexander Denecke, Heiko Wersing, Jochen J. Steil,...
This paper describes techniques for fusing the output of multiple cues to robustly and accurately segment foreground objects from the background in image sequences. Two different m...
Global shape information is an effective top-down complement
to bottom-up figure-ground segmentation as well
as a useful constraint to avoid drift during adaptive tracking.
We p...
A geometric construction is presented which is shown to be an effective tool for understanding and implementing multi-category support vector classification. It is demonstrated ho...
The problem of hypertext classification deals with objects possessing more complex information structure than the plain text has. Present hypertext classification systems show the...