Multi-operator retargeting based on perceptual structural similarity

We propose a new multi-operator retargeting algorithm by using three resizing operators of seam carving, scaling, and cropping iteratively. To determine which operator should be used at each iteration, we adopt structural similarity (SSIM) to evaluate the similarity between the original and retargeted images for the dynamic programming. Since the sizes of original and retargeted images are different, SIFT flow is used for dense correspondence between the original and retargeted images for similarity evaluation. Additionally, visual saliency is used to weight SSIM results based on the characteristics of the Human Visual System (HVS). Experimental results on a public image retargeting database show the promising performance of the proposed multi-operator retargeting algorithm.