D$^{2}$C-Net: A Dual-Branch, Dual-Guidance and Cross-Refine Network for Camouflaged Object Detection

In this article, we propose a novel framework for camouflaged object detection (COD), named D $^{2}$ C-Net, which contains two new modules: Dual-branch features extraction (DFE) and gradually refined cross fusion (GRCF). Specifically, the DFE simulates two-stage process of human visual mechanisms in observing camouflage scenes. For first stage, dense concatenation is employed to aggregate multilevel expand receptive field. The stage feature maps are then utilized extract two-direction guidance information, benefits second stage. GRCF consists self-refine attention unit cross-refinement unit, with aim combining peer layer an improved COD performance. proposed outperforms 13 state-of-the-art deep learning-based methods upon three public datasets terms five widely used metrics. Finally, show evidence successful applications method fields surface defect medical image segmentation.