On the matching of images containing repetitive patterns by pairs of interest points

Image matching is a fundamental task for computer vision. However, due to the local ambiguities induced by repetitive patterns in most of man-made objects or scenes, matching images of repetitive patterns is still a challenging problem even the viewpoint changes of them are not very large. This paper is primarily focused on the problem of matching images containing repetitive patterns and proposes a novel method based on pairs of interest points to solve the problem. It starts from matching pairs of interest points and then obtaining point correspondences from the matched point-pairs based on the low distortion constraint, which is meant that the distortions of point groups should be small across images. By combing pairs of interest points, local ambiguities induced by repetitive patterns can be reduced to some extent since information in a much larger region is used. Moreover, owing to our newly defined compatibility measure between one correspondence and a set of point correspondences, the obtained point correspondences are very reliable. Experimental results have demonstrated the effectiveness of the proposed method as well as its superiority to the existing methods.