Double cross-modality progressively guided network for RGB-D salient object detection

作者：

Highlights：

• Design a double dilated merge module to extract cross-modality features.

• Propose a cross-modality enhance module to enhance cross-modalities compatibility.

• Introduce a distribution-region combination loss.

摘要

Highlights•Design a double dilated merge module to extract cross-modality features.•Propose a cross-modality enhance module to enhance cross-modalities compatibility.•Introduce a distribution-region combination loss.

论文关键词：RGB-D,Salient object detection,Cross-modality,Attention mechanism,Integration

论文评审过程：Received 8 May 2021, Revised 25 October 2021, Accepted 24 November 2021, Available online 29 November 2021, Version of Record 15 December 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104351