FCL-COD: Weakly Supervised Camouflaged Object Detection with Frequency-aware and Contrastive Learning
This work addresses the challenge of reducing annotation costs for camouflaged object detection, which is important for applications like surveillance and biology, but it is incremental as it builds on existing models like SAM.
The paper tackles the problem of weakly supervised camouflaged object detection (WSCOD) by proposing FCL-COD, a framework that integrates frequency-aware and contrastive learning to address issues like non-camouflage responses and poor boundaries, achieving results that surpass state-of-the-art weakly and fully supervised methods on three benchmarks.
Existing camouflage object detection (COD) methods typically rely on fully-supervised learning guided by mask annotations. However, obtaining mask annotations is time-consuming and labor-intensive. Compared to fully-supervised methods, existing weakly-supervised COD methods exhibit significantly poorer performance. Even for the Segment Anything Model (SAM), there are still challenges in handling weakly-supervised camouflage object detection (WSCOD), such as: a. non-camouflage target responses, b. local responses, c. extreme responses, and d. lack of refined boundary awareness, which leads to unsatisfactory results in camouflage scenes. To alleviate these issues, we propose a frequency-aware and contrastive learning-based WSCOD framework in this paper, named FCL-COD. To mitigate the problem of non-camouflaged object responses, we propose the Frequency-aware Low-rank Adaptation (FoRA) method, which incorporates frequency-aware camouflage scene knowledge into SAM. To overcome the challenges of local and extreme responses, we introduce a gradient-aware contrastive learning approach that effectively delineates precise foreground-background boundaries. Additionally, to address the lack of refined boundary perception, we present a multi-scale frequency-aware representation learning strategy that facilitates the modeling of more refined boundaries. We validate the effectiveness of our approach through extensive empirical experiments on three widely recognized COD benchmarks. The results confirm that our method surpasses both state-of-the-art weakly supervised and even fully supervised techniques.