Boundary-guided network for camouflaged object detection

作者：

Highlights：

• A real-time high-performance method is proposed for camouflaged object detection.

• Using global context information and local details to boost the performance.

• Camouflaged regions and their boundaries are combined to boost the performance.

• The proposed method outperforms other state-of-the-art models on four datasets.

摘要

•A real-time high-performance method is proposed for camouflaged object detection.•Using global context information and local details to boost the performance.•Camouflaged regions and their boundaries are combined to boost the performance.•The proposed method outperforms other state-of-the-art models on four datasets.

论文关键词：Boundary guidance,Camouflage object detection,Convolutional neural network,Coarse-to-fine refinement

论文评审过程：Received 20 December 2021, Revised 31 March 2022, Accepted 22 April 2022, Available online 30 April 2022, Version of Record 13 May 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.108901