Indexing based on scale invariant interest points

TLDR

Scale‑space theory shows that interest points can be adapted to scale for repeatable, geometrically stable detection, and that local extrema of normalized derivatives across scales reveal characteristic local structures. The paper proposes a new method for detecting scale‑invariant interest points. The method builds a multi‑scale Harris representation, selects points where the Laplacian peaks across scales, and uses the associated scale to compute a scale‑invariant descriptor for indexing. The approach yields distinctive, scale‑invariant points that are robust to scale, rotation, translation, illumination, and limited viewpoint changes, and the resulting descriptors achieve excellent indexing performance up to a scale factor of four on a database of over 5,000 images.

Abstract

This paper presents a new method for detecting scale invariant interest points. The method is based on two recent results on scale space: (1) Interest points can be adapted to scale and give repeatable results (geometrically stable). [2] Local extrema over scale of normalized derivatives indicate the presence of characteristic local structures. Our method first computes a multi-scale representation for the Harris interest point detector. We then select points at which a local measure (the Laplacian) is maximal over scales. This allows a selection of distinctive points for which the characteristic scale is known. These points are invariant to scale, rotation and translation as well as robust to illumination changes and limited changes of viewpoint. For indexing, the image is characterized by a set of scale invariant points; the scale associated with each point allows the computation of a scale invariant descriptor. Our descriptors are, in addition, invariant to image rotation, of affine illumination changes and robust to small perspective deformations. Experimental results for indexing show an excellent performance up to a scale factor of 4 for a database with more than 5000 images.