Image retargeting quality assessment based on saliency-driven classification

作者：

Highlights：

• We propose a novel salient region extraction method for retargeted images.

• We develop an image classification algorithm using the foreground and gradient maps.

• We design a salient region-based ARS metric to measure the geometric distortion.

摘要

•We propose a novel salient region extraction method for retargeted images.•We develop an image classification algorithm using the foreground and gradient maps.•We design a salient region-based ARS metric to measure the geometric distortion.

论文关键词：Image retargeting quality assessment,Aspect ratio similarity,Image classification,Backward registration,Salient region

论文评审过程：Received 27 August 2020, Revised 6 March 2022, Accepted 14 March 2022, Available online 18 March 2022, Version of Record 2 April 2022.

论文官网地址：https://doi.org/10.1016/j.image.2022.116688