Image spatial verification using segment intersection of interest points

Abstract

This work presents a new spatial verification technique for image similarity search. The proposed algorithm evaluates the geometry of the detected local keypoints by building segments connecting pairs of points and analyzing their intersections in a 2D plane. We show that these intersections remain constant with respect to different geometric transformations (translation, rotation, similarity and affine). Evaluation has been performed obtaining an initial image similarity ranking with a BOF-based methodology, and then using the proposed method for reranking. The presented algorithm (SIIP) has been compared to the RANSAC spatial verification method, showing that it is more efficient and obtains a higher performance on three different datasets.