Rethinking the U-Shape Structure for Salient Object Detection

The U-shape structure has shown its advantage in salient object detection for efficiently combining multi-scale features. However, most existing U-shape-based methods focused on improving the bottom-up and top-down pathways while ignoring the connections between them. This paper shows that we can achieve the cross-scale information interaction by centralizing these connections, hence obtaining semantically stronger and positionally more precise features. To inspire the newly proposed strategy's potential, we further design a relative global calibration module that can simultaneously process multi-scale inputs without spatial interpolation. Our approach can aggregate features more effectively while introducing only a few additional parameters. Our approach can cooperate with various existing U-shape-based salient object detection methods by substituting the connections between the bottom-up and top-down pathways. Experimental results demonstrate that our proposed approach performs favorably against the previous state-of-the-arts on five widely used benchmarks with less computational complexity. The source code will be publicly available.