An improved 3D object feature points correspondence algorithm

The current approaches to feature points´ correspondence are based on the hypothesis that the displacements for feature points between consecutive frames are small. That is, there is very short time interval between successive two frames in image sequences. These approaches are difficult to find the corresponding feature points when the time interval is large and the 3D object moves in the space. Someone proposed a 3D object feature points´ correspondence. The first step is evaluating the object´s pose by matching the input image and the reference image in the object model database. But the object model database has so many images, so it will spend much time to match. An improved method with quick search object model database algorithm is proposed in this paper to resolve this problem. It is turned out with experiment that the method presented in this paper is an effective method to 3D object feature points´ correspondence.