Image matching based on a local invariant descriptor

Image matching is a fundamental task of many computer vision problems. In this paper we present a novel approach to match two images in presenting significant geometric deformations and considerable photometric variations. The approach is based on local invariant features. First, local invariant regions are detected by a three-step process which determines the positions, scales and orientations of the regions. Then each region is represented by a novel descriptor. The descriptor is a two-dimensional histogram. Performance evaluations show that this new descriptor generally provides higher distinctiveness and robustness to image deformations. We present the image matching results. The matching results show good performance of our approach for both geometric deformations and photometric variations.