Integrating kAS and SIFT-like descriptor for image description

Shape-descriptor (e.g. Adjacent Contour Segments, i.e. kAS) and keypoint-descriptor (e.g. Scale Invariant Feature Transform, i.e. SIFT) are widely used for computer vision. However, few works principally integrate shape-descriptor and keypoint-descriptor to describe the content of an image. On one hand, in some cases the degree of locality of keypiont-descriptor is too high to capture semantic characteristics of an object. On the other hand, though the shape has higher semantic level than keypoint, it contains no texture information because only the information of contour/edge is used. To make full use of the information of both shape and keypoint for generate robust and distinctive features, in this paper we propose an algorithm to integrate shape and keypoint descriptor. Specifically, we employ kAS to extract useful shape information. Then keypoints of a kAS shape are defined at which we propose to extract SIFT-like features. Experimental results on image matching demonstrate the effectiveness of the proposed algorithm.