Search for a command to run...
Addresses the problem of similar-shape retrieval, where shapes or images in a shape database that satisfy specified shape-similarity constraints with respect to the query shape or image must be retrieved from the database. In its simplest form, the similar-shape retrieval problem can be stated as, "retrieve or select all shapes or images that are visually similar to the query shape or the query image's shape". We focus on databases of 2D shapes-or equivalently, databases of images of flat or almost flat objects. (We use the terms "object" and "shape" interchangeably). Two common types of 2D objects are rigid objects, which have a single rigid component called a link, and articulated objects, which have two or more rigid components joined by movable (rotating or sliding) joints. An ideal similar-shape retrieval technique must be general enough to handle images of articulated as well as rigid objects. It must be flexible enough to handle simple query images, which have isolated shapes, and complex query images, which have partially visible, overlapping or touching objects. We discuss the central issues in similar-shape retrieval and explain how these issues are resolved in a shape retrieval scheme called FIBSSR (Feature Index-Based Similar-Shape Retrieval). This new similar-shape retrieval system effectively models real-world applications.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>