Multi-scale feature aggregation and boundary awareness network for salient object detection

作者：

Highlights：

• A multi-scale feature aggregation and boundary awareness network for salient object detection is proposed.

• We adopt a multiple aggregation strategy to reduce the loss of local details in multi-level feature fusion.

• A boundary pixel awareness loss is designed to solve complex bondary problem.

摘要

•A multi-scale feature aggregation and boundary awareness network for salient object detection is proposed.•We adopt a multiple aggregation strategy to reduce the loss of local details in multi-level feature fusion.•A boundary pixel awareness loss is designed to solve complex bondary problem.

论文关键词：Deep learning,Salient object detection,Feature aggregation,Complex boundary

论文评审过程：Received 16 February 2022, Accepted 18 March 2022, Available online 26 March 2022, Version of Record 11 April 2022.

论文官网地址：https://doi.org/10.1016/j.imavis.2022.104442