A. Petrosino (Ed.): ICIAP 2013, Part II, LNCS 8157, pp. 151–160, 2013. © Springer-Verlag Berlin Heidelberg 2013 Saliency Based Aesthetic Cut of Digital Images Luca Greco and Marco La Cascia DICGIM, Università degli Studi di Palermo, Italy {luca.greco,marco.lacascia}@unipa.it Abstract. Aesthetic cut of photos is a process well known to professional photographers. It consists of cutting the original photo to remove less relevant parts close to the borders leaving in this way the interesting subjects in a position that is perceived by the observer as more pleasant. In this paper we propose a saliency based technique to automatically perform aesthetic cut in images. We use a standard method to estimate the saliency map and propose some post processing on the map to make it more suitable for our scope. We then apply a greedy algorithm to determine the cut (i.e. the most important part of the original image) both in the cases of free and fixed aspect ratio. Experimental results are reported showing how the cut resulting from our technique compares to some state of the art retargeting and cropping techniques. Keywords: Aesthetic cut, image editing, saliency. 1 Introduction and Related Works Image saliency refers to those elements that a human observer consider important or attracting his attention. Sometime analyzing the visibility and the most visible elements of an image is also used the term conspicuity. An inner part of the image is important if it is different from the rest of other regions, gaining greater visibility and attracting attention, resulting in catching the eyes focus. Salient parts of a scene evoke a strong visual response and polarize attention, detaching this zones from background. Human visual attention is composed of two factors connected with two stimuli of different nature: • the objective factor depends exclusively on the characteristics of the image; • the subjective factor is related to the subject's will. The first has a bottom-up activation, and is related to physical characteristics such as brightness, color and shape. The second has a top-down activation and is influenced by the knowledge obtained by learning the probabilistic structure of the environment. In many situations the second give the greatest contribution to the acquisition of information. Saliency can be used in Computer Vision tasks for smart thumbnailing or subject extraction. In this paper saliency map is used to get an aesthetic cut of pictures.