Content-Aware Image Resizing Based on Row/Column Clustering

Abstract

We propose an image resizing method using row/column removal and addition based on the clustering of rows and columns in an image. To shrink an image, we first separately cluster a set of rows and a set of columns in the original image and then replace each cluster with a representative row or column to reduce the numbers of rows and columns. To enlarge an image, we add linearly interpolated rows and columns between adjacent clusters. Experimental results show that the proposed method can shrink or enlarge given images while preserving the aspect ratios of the main objects in the images.