Salient object detection employing robust sparse representation and local consistency

作者：

Highlights：

• Salient object detection is formulated as a robust sparse pursuit problem via RSR.

• A local consistency prior is integrated with RSR to uniformly highlight the whole object.

• Representation coefficients and sparse errors jointly determine the saliency measure.

• The method works well for images with complex structures.

摘要

•Salient object detection is formulated as a robust sparse pursuit problem via RSR.•A local consistency prior is integrated with RSR to uniformly highlight the whole object.•Representation coefficients and sparse errors jointly determine the saliency measure.•The method works well for images with complex structures.

论文关键词：Salient object detection,Robust sparse representation,Local consistency,Complex structures

论文评审过程：Received 28 December 2016, Revised 17 August 2017, Accepted 20 October 2017, Available online 28 October 2017, Version of Record 6 February 2018.

论文官网地址：https://doi.org/10.1016/j.imavis.2017.10.002