Local grayvalue invariants for image retrieval

TLDR

The paper proposes a method for retrieving images from large databases. It uses local grayvalue invariants computed at automatically detected interest points, combined with a voting algorithm and semilocal constraints to enable retrieval. The approach indexes over 1,000 images and successfully retrieves correct matches under partial visibility, similarity transformations, extraneous features, and minor perspective deformations.

Abstract

This paper addresses the problem of retrieving images from large image databases. The method is based on local grayvalue invariants which are computed at automatically detected interest points. A voting algorithm and semilocal constraints make retrieval possible. Indexing allows for efficient retrieval from a database of more than 1,000 images. Experimental results show correct retrieval in the case of partial visibility, similarity transformations, extraneous features, and small perspective deformations.