Generation of Photomosaic with Saliency Map

Abstract

Photomosaic is a technique for representing a single image by arranging small images in a grid. In general, the input image is divided into a grid and a small image whose average color is closest to the average color of grid is chosen, however, important features such as contours can be lost. In this paper, we propose a method for creating photomosaics with saliency map to take shape features in the input image into account.