Title:CoFiNet: Unveiling Camouflaged Objects with Multi-Scale Finesse

Abstract:Camouflaged Object Detection (COD) is a critical aspect of computer vision aimed at identifying concealed objects, with applications spanning military, industrial, medical and monitoring domains. To address the problem of poor detail segmentation effect, we introduce a novel method for camouflage object detection, named CoFiNet. Our approach primarily focuses on multi-scale feature fusion and extraction, with special attention to the model's segmentation effectiveness for detailed features, enhancing its ability to effectively detect camouflaged objects. CoFiNet adopts a coarse-to-fine strategy. A multi-scale feature integration module is laveraged to enhance the model's capability of fusing context feature. A multi-activation selective kernel module is leveraged to grant the model the ability to autonomously alter its receptive field, enabling it to selectively choose an appropriate receptive field for camouflaged objects of different sizes. During mask generation, we employ the dual-mask strategy for image segmentation, separating the reconstruction of coarse and fine masks, which significantly enhances the model's learning capacity for details. Comprehensive experiments were conducted on four different datasets, demonstrating that CoFiNet achieves state-of-the-art performance across all datasets. The experiment results of CoFiNet underscore its effectiveness in camouflage object detection and highlight its potential in various practical application scenarios.