Improved salient object detection using hybrid Convolution Recurrent Neural Network

作者：

Highlights：

• A new approach is proposed to find the salient objects from a video dataset.

• It uses CRNN to captures temporal, spatial, and local constraint features.

• It evaluated on benchmark datasets against existing video salient object detection.

• The experiment shows it can substantially outperform other existing saliency models.

摘要

•A new approach is proposed to find the salient objects from a video dataset.•It uses CRNN to captures temporal, spatial, and local constraint features.•It evaluated on benchmark datasets against existing video salient object detection.•The experiment shows it can substantially outperform other existing saliency models.

论文关键词：Convolution Recurrent Neural Network,Salient object detection,Spatial features,Temporal features,Local constraint cues

论文评审过程：Received 20 June 2019, Revised 22 August 2020, Accepted 26 September 2020, Available online 1 October 2020, Version of Record 14 October 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114064