Fast Camouflaged Object Detection via Edge-based Reversible Re-calibration Network

Camouflaged Object Detection (COD) aims to detect objects with similar patterns (e.g., texture, intensity, colour, etc) their surroundings, and recently has attracted growing research interest. As camouflaged often present very ambiguous boundaries, how determine object locations as well weak boundaries is challenging also the key this task. Inspired by biological visual perception process when a human observer discovers objects, paper proposes novel edge-based reversible re-calibration network called ERRNet. Our model characterized two innovative designs, namely Selective Edge Aggregation (SEA) Reversible Re-calibration Unit (RRU), which aim behaviour achieve effective edge prior cross-comparison between potential regions background. More importantly, RRU incorporates diverse priors more comprehensive information comparing existing COD models. Experimental results show that ERRNet outperforms cutting-edge baselines on three datasets five medical image segmentation datasets. Especially, compared top-1 SINet, significantly improves performance ?6% (mean E-measure) notably high speed (79.3 FPS), showing could be general robust solution for