Visual saliency detection using information divergence

作者：

Highlights：

• A novel and efficient framework for saliency detection is proposed.

• Independent component analysis makes the model flexible for extracting features.

• Visual saliency is interpreted by the difference between information in image.

• The model achieves state-of-the-art on large-scale natural image database.

摘要

•A novel and efficient framework for saliency detection is proposed.•Independent component analysis makes the model flexible for extracting features.•Visual saliency is interpreted by the difference between information in image.•The model achieves state-of-the-art on large-scale natural image database.

论文关键词：Visual attention,Saliency detection,Independent component analysis,Bayesian surprise model

论文评审过程：Received 14 March 2012, Revised 7 March 2013, Accepted 9 March 2013, Available online 1 April 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.03.008