An Image Mosaic Algorithm for Spherical Panorama

Visual reality is one of the hot topics in these years. Panoramic photography is one of the VR technique. Image mosaic algorithms is the main technology to stitch the little scale images to a large scale panoramic image. In this paper, an improved image mosaic method is presented. Because of using fish-eye images, spherical panorama model was chosen to describe the alignment between a pair of images. A feature-based method is devised to estimate the parameters of the registration model. In order to reduce registration errors, optical flow was applied to local alignment. Before blending the images, a brightness adjustment algorithm base on sample point will be introduced.