Affine-invariant image retrieval by correspondence matching of shapes

In this paper, we present an efficient multi-scale similarity matching meth od for shape-based image indexing and retrieval. This method is affine-inva riant and stable against noise and shape deformations. Shapes which have un dergone mirror reflection can also be retrieved in a unified manner. In thi s approach, similarity matching is cast as correspondence matching of two s hapes which is then solved by minimizing the matching errors between two fe ature vectors. Since our feature vectors simultaneously capture both local and global affine-invariant features of shapes, this formulation makes our solution to the correspondence problem very robust. To render the technique suitable for interactive image retrieval, a fast error minimization algori thm for computing correspondence matching is further proposed. Theoretical analysis and experimental results show that multi-scale similarity matching allows dissimilar shapes to be filtered out very quickly and the resulting method meets the performance and flexibility needed for content-based imag e indexing and retrieval. (C) 1999 Elsevier Science B.V. All rights reserve d.