Weakly supervised camouflaged object detection (WS-COD) aims to address the critical task of identifying visually assimilated objects concealed within heterogeneous backgrounds under sparse supervisory signals. However, current WS-COD frameworks suffer from compromised structural integrity, stemming from cross-hierarchical feature discrepancy and constrained cross-level information flow, which induces structural misalignment and context fragmentation in multi-granularity feature fusion. To overcome the limitation, we propose a novel SAM-guided Resolution Iteration Learning Network (SAM-RNet) that synergizes foundation model priors with multi-resolution feature refinement. Our technical contributions are threefold: (1) We utilize the Segment Anything Model (SAM) to produce high-quality masks, effectively mitigating supervision insufficiency through large-scale visual knowledge distillation. (2) We design a resolution iteration mechanism where high-resolution features progressively refine low-resolution counterparts through an Interactive Refinement Module (IRM) - a dual-branch architecture enabling hierarchical feature interaction and enhancement through branch collaboration and attention mechanism, complemented by an iterative feedback loss to enforce multi-scale feature learning. (3) We develop a Decoder with cross-layer fusion operations, enabling the aggregation of features from object and background contexts for precise object segmentation. Finally, extensive experiments demonstrate that SAM-RNet is superior to existing WS-COD methods across three COD datasets, achieving average improvements of 4.37%, 4.60%, 7.00%, and 24.06% in$S_{\alpha }$,$E_{\phi }$,$F_{\beta }^{\omega }$, and$M$, respectively.

The purpose of co-salient object detection (CoSOD) is to detect the salient objects that co-occur in a group of relevant images. CoSOD has been significantly prospered by recent advances in convolutional neural networks (CNNs). However, it shows general limitations in modeling long-range feature dependencies, which is crucial for CoSOD. In the vision transformer, the self-attention mechanism is utilized to capture global dependencies but unfortunately destroy local spatial details, which are also essential for CoSOD. To address the above issues, we propose a dual network structure, called TCNet, which can efficiently excavate both local information and global representations for co-saliency learning via the parallel interaction of Transformers and CNNs. Specifically, it contains three critical components, i.e., the mutual consensus module (MCM), the consensus complementary module (CCM), and the group consistent progressive decoder (GCPD). MCM aims to capture the global consensus from high-level features of these two branches as a guide for the following integration of consensus cues of both branches at each level. Next, CCM is designed to effectively fuse the consensus of local information and global contexts from different levels of the two branches. Finally, GCPD is developed to maintain group feature consistency and predict accurate co-saliency maps. The proposed TCNet is evaluated on five challenging CoSOD benchmark datasets using six widely used metrics, showing that our proposed method is superior to other existing cutting-edge methods for co-salient object detection.

Salient object detection is used as a pre-process in many computer vision tasks (such as salient object segmentation, video salient object detection, etc.). When performing salient object detection, depth information can provide clues to the location of target objects, so effective fusion of RGB and depth feature information is important. In this paper, we propose a new feature information aggregation approach, weighted group integration (WGI), to effectively integrate RGB and depth feature information. We use a dual-branch structure to slice the input RGB image and depth map separately and then merge the results separately by concatenation. As grouped features may lose global information about the target object, we also make use of the idea of residual learning, taking the features captured by the original fusion method as supplementary information to ensure both accuracy and completeness of the fused information. Experiments on five datasets show that our model performs better than typical existing approaches for four evaluation metrics.