Image saliency prediction by learning deep probability model

作者：

Highlights：

• An active ranking algorithm is proposed for salient objects ranking.

• A hierarchy CNNs architecture is utilized to extract deep representation of salient regions.

• A probabilistic model is learned to guide image saliency prediction.

摘要

•An active ranking algorithm is proposed for salient objects ranking.•A hierarchy CNNs architecture is utilized to extract deep representation of salient regions.•A probabilistic model is learned to guide image saliency prediction.

论文关键词：Image saliency prediction,Probability model,Deeply-learned feature

论文评审过程：Received 20 February 2019, Revised 19 July 2019, Accepted 1 August 2019, Available online 2 August 2019, Version of Record 20 August 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.08.002