Camouflaged Object Detection with Adaptive Partition and Background Retrieval

Abstract

Recent works confirm the importance of local details for identifying camouflaged objects. However, how to identify the details around the target objects via background cues lacks in-depth study. In this paper, we take this into account and present a novel learning framework for camouflaged object detection, called AdaptCOD. To be specific, our method decouples the detection process into three parts, namely localization, segmentation, and retrieval. We design a context adaptive partition strategy to dynamically select a reasonable context region for local segmentation and a background retrieval module to further polish the camouflaged object boundaries. Despite the simplicity, our method enables even a simple COD model to achieve great performance. Extensive experiments show that AdaptCOD surpasses all existing state-of-the-art methods on three widely-used camouflaged object detection benchmarks. Code is publicly available at https://github.com/HVision-NKU/AdaptCOD.