I. Introduction
Image/Video stitching focuses on generating wide-view, high-resolution contents from overlapping images/videos. It has been widely applied on various applications such as aerospace, automobiles, security monitoring, surveillance, virtual reality, sports broadcasting, and so on [1]. Numerous approaches have been reported in the literature, which can be categorized into three main groups, i.e., image, video, and panoramic stitching. It involves three major steps for performing image/video stitching, including (a) feature extraction and description, (b) keypoint matching, (c) transforming the matching points to a single coordinate system.