Attention Retargeting by Color Manipulation in Images
Victor A. Mateescu, Ivan V. Bajić · 2014
Attention retargeting in images is a concept in which the content or composition of the image is altered in an effort to guide the viewer's attention to a specific location. In this paper, we propose a method that modifies the color of a selected region in an image to increase its saliency and draw attention towards it. To avoid many of the issues present in existing approaches to attention retargeting, including high computational complexity and unnatural-looking modifications to the images, we make the case for adjusting hue while leaving all remaining color components fixed. By representing the hue as an angle in CIE L*a*b* color space, we may express an adjustment in hue as a rotation in this space. The optimal hue adjustment is the rotation that maximizes the dissimilarity of hue distribution of the selected region relative to its surroundings. We apply our method on a set of natural images and confirm its effectiveness in guiding attention through eye-tracking. The naturalness of the results are evaluated in a separate set of subjective experiments.