SAM-COD: SAM-guided Unified Framework for Weakly-Supervised Camouflaged Object Detection

TL;DR

SAM-COD introduces a unified weakly-supervised framework for camouflaged object detection that effectively supports various label types and improves mask quality using prompt adaptation and knowledge distillation, outperforming existing methods.

Contribution

It proposes a novel unified framework supporting arbitrary weakly-supervised labels for camouflaged object detection, integrating prompt adaptation, response filtering, semantic matching, and knowledge distillation.

Findings

Outperforms state-of-the-art weakly-supervised methods

Achieves comparable or superior results to fully-supervised approaches

Validated on three mainstream COD benchmarks

Abstract

Most Camouflaged Object Detection (COD) methods heavily rely on mask annotations, which are time-consuming and labor-intensive to acquire. Existing weakly-supervised COD approaches exhibit significantly inferior performance compared to fully-supervised methods and struggle to simultaneously support all the existing types of camouflaged object labels, including scribbles, bounding boxes, and points. Even for Segment Anything Model (SAM), it is still problematic to handle the weakly-supervised COD and it typically encounters challenges of prompt compatibility of the scribble labels, extreme response, semantically erroneous response, and unstable feature representations, producing unsatisfactory results in camouflaged scenes. To mitigate these issues, we propose a unified COD framework in this paper, termed SAM-COD, which is capable of supporting arbitrary weakly-supervised labels. Our SAM-COD employs a prompt adapter to handle scribbles as prompts based on SAM. Meanwhile, we introduce response filter and semantic matcher modules to improve the quality of the masks obtained by SAM under COD prompts. To alleviate the negative impacts of inaccurate mask predictions, a new strategy of prompt-adaptive knowledge distillation is utilized to ensure a reliable feature representation. To validate the effectiveness of our approach, we have conducted extensive empirical experiments on three mainstream COD benchmarks. The results demonstrate the superiority of our method against state-of-the-art weakly-supervised and even fully-supervised methods.