Dense Attention Fluid Network for Salient Object Detection in Optical Remote Sensing Images

Despite the remarkable advances in visual saliency analysis for natural scene images (NSIs), salient object detection (SOD) optical remote sensing (RSIs) still remains an open and challenging problem. In this paper, we propose end-to-end Dense Attention Fluid Network (DAFNet) SOD RSIs. A Global Context-aware (GCA) module is proposed to adaptively capture long-range semantic context relationships, further embedded a (DAF) structure that enables shallow attention cues flow into deep layers guide generation of high-level feature maps. Specifically, GCA composed two key components, where global aggregation achieves mutual reinforcement embeddings from any spatial locations, cascaded pyramid tackles scale variation issue by building up framework progressively refine map coarse-to-fine manner. addition, construct new RSI dataset contains 2,000 with pixel-wise annotations, which currently largest publicly available benchmark. Extensive experiments demonstrate our DAFNet significantly outperforms existing state-of-the-art competitors. https://github.com/rmcong/DAFNet_TIP20