Lowcomplexity Content-Aware Image Retargeting

Kairan Sun,Bo Yan,Yiqi Gao
DOI: https://doi.org/10.1109/icip.2012.6467307
2012-01-01
Abstract:Image retargeting plays a more and more significant role recently, thanks to the escalating diversity of display devices. In this paper, we present a novel low complexity content-aware image retargeting method, which can provide both high efficiency and quality. Specifically, the importance map is used as the basis of our method. The important regions tend to preserve its original size. Our method performs inter-row coherence filtering on importance maps in order to maintain the spatial coherence, and then directly utilizes the filtered importance maps to generate the scaling map. Experimental results show the proposed algorithm improves the efficiency significantly compared with other existing methods. At the same time, the resized image quality of our algorithm is as good as, if not better than, that of the other methods. As a result, our method possesses huge practical significance.
What problem does this paper attempt to address?