Adaptive Segmentation and Stitching on 8K UHD Video

Abstract

Transmission technology is necessary to display contents stored on the Cloud server. When excessive compression is performed to the transmission of ultra-high resolution image, tangibility is reduced. Therefore, in this paper, we proposed an algorithm that divides the image into a number of sub-images. The sub-images were restored to the original one by stitching at the receiver. In the existing study, important objects were located at the center of the images, but an exception occurred when they are at the edge of the images. The saliency map was used to detect their main part so that the region of interest will not be divided. The images were divided depending on the position of saliency map. It is expected that users will be able to provide realistic signage by displaying ultra-high resolution images with a large screen.