In this paper a functional approach to object description for computer vision is suggested as a tool to deal efficiently with the wide variations of the shapes of many common objects. Two main problems are discussed. The first one is how to build functional descriptions, starting from a set of functional primitives, by means of a suitable conceptual syntax. The second one is how to relate functional descriptions to the visual data; this involves both the management of the interactions among several experts and the definition of triggering features to avoid extensive searches.
Building functional descriptions
Giunchiglia, Fausto;
1985-01-01
Abstract
In this paper a functional approach to object description for computer vision is suggested as a tool to deal efficiently with the wide variations of the shapes of many common objects. Two main problems are discussed. The first one is how to build functional descriptions, starting from a set of functional primitives, by means of a suitable conceptual syntax. The second one is how to relate functional descriptions to the visual data; this involves both the management of the interactions among several experts and the definition of triggering features to avoid extensive searches.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione
