Correspondence Estimation between Objects using Image Matching Techniques

Abstract

Image Matching is the process of bringing two images geometrically into agreement that is the corresponding pixels in the two images corresponds to the same physical object. It establishes correspondence between each pair of visible homologous image parts for the considered pair of images and aims towards identification of similar objects. In this paper, SIFT and ORB algorithm is used for identifying the similar objects among different scenes. It can be used to identify similar objects from different scenes and to identify the similarity between given two objects that are in different rotation orientations and different backgrounds.