Hedging static saliency models to predict dynamic saliency

作者：

Highlights：

• We propose a novel weakly supervised dynamic saliency model called HedgeSal.

• Our model depends on a decision-theoretic online learning algorithm.

• We process appearance and motion streams via pre-trained deep static saliency models.

• Combining their decisions in an adaptive manner yields state-of-the-art results.

摘要

•We propose a novel weakly supervised dynamic saliency model called HedgeSal.•Our model depends on a decision-theoretic online learning algorithm.•We process appearance and motion streams via pre-trained deep static saliency models.•Combining their decisions in an adaptive manner yields state-of-the-art results.

论文关键词：Dynamic saliency,Hedge algorithm,Decision theoretic online learning,Feature integration

论文评审过程：Received 27 December 2018, Revised 6 November 2019, Accepted 6 November 2019, Available online 13 November 2019, Version of Record 14 November 2019.

论文官网地址：https://doi.org/10.1016/j.image.2019.115694