检索结果-内蒙古大学图书馆

camouflaged object detection with Adaptive Partition and Background Retrieval

INTERNATIONAL JOURNAL OF COMPUTER VISION 2025年 1-17页

作者： Yin, Bowen Zhang, Xuying Liu, Li Cheng, Ming-Ming Liu, Yongxiang Hou, Qibin Nankai Univ VCIP CS Tianjin Peoples R China NUDT Coll Elect Sci & Technol Changsha Peoples R China

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://***/HVision-NKU/AdaptCOD.

关键词： camouflaged object detection Adaptive partition Background retrieval

来源：评论

学校读者我要写书评

暂无评论

camouflaged object detection via boundary refinement

引用

MULTIMEDIA SYSTEMS 2025年第1期31卷 1-20页

作者： Zhang, Miaohui Shen, Chenxing Deng, Yangyang Wang, Li Henan Univ Sch Artificial Intelligence Zhengzhou 450046 Peoples R China Henan Univ Eurasia Int Sch Kaifeng 475001 Peoples R China

In camouflaged object detection (COD), wholly and accurately segmenting the foreground from the background is a major focus of research. However, the similarity in color and texture between foreground targets and the background makes it difficult to distinguish them. Despite numerous deep learning networks utilizing different approaches for camouflaged object detection, precisely identifying object contours and enhancing the model's resistance to interference from similar backgrounds remains a challenging problem. To address this issue, this paper proposes a camouflaged object detection network named boundary refinement network (BRNet) that achieves a fine-grained description of object contours by utilizing boundary semantic information constraints. Firstly, the multi-level asymmetric convolution module (MACM) is designed to enhance feature representation within the backbone architecture via a sequence of asymmetric convolutions and cross-layer connections. Additionally, the Boundary Constraint Guided Module (BCGM) is proposed to impose constraints on foreground shape and refine constrained foreground contours. Lastly, we introduce the Boundary Fusion Extraction Module (BFEM), which enables interaction between boundaries and objects in an additional dimension, leading to the generation of prediction results. Extensive quantitative and qualitative experiments conducted on three datasets demonstrate that BRNet performs well on the camouflaged object detection task, achieving superior results compared to 21 state-of-the-art approaches.

关键词： camouflaged object detection Boundary constraint Boundary refinement 3D convolution

来源：评论

学校读者我要写书评

暂无评论

camouflaged object detection Based on Feature Aggregation and Global Semantic Learning 7th

Camouflaged Object Detection Based on Feature Aggregation an...

引用

7th Chinese Conference on Pattern Recognition and Computer Vision

作者： Wang, Kuan Li, Xiuhong Li, Boyuan Li, Songlin Wei, Zijun Wan, Lining Xinjiang Univ Sch Informat Sci & Engn Urumqi Xinjiang Peoples R China Xinjiang Key Lab Signal Detect & Proc Urumqi Xinjiang Peoples R China

ISBN: (纸本)9789819788576;9789819788583

In camouflage object detection (COD), a large amount of contextual information is usually required because the object is very similar to its surrounding environment. However, due to the current problems of insufficient feature aggregation in the backbone network and insufficient effective interaction between high and low semantics, it is difficult to fully mine contextual information. This situation hampers the differentiation between background and targets, leading to potential misjudgments. To address these issues, we propose a feature aggregation and global semantic learning network (FAGSL-Net). We design a Feature Aggregation Enhancement Module (FAEM) to aggregate features from different levels, which progressively aggregates feature maps from high to low levels in a cascaded manner to obtain robust fused feature representations, further generating preliminary forecast chart. Then, leveraging the designed Global Semantic Fusion Module (GSFM), we perform high and low semantic interaction learning between the preliminary forecast chart containing multi-level features and low-level feature maps to comprehensively utilize semantic information from different levels to enhance camouflage object detection performance. Finally, we utilize a multi-scale strategy to mine information from global features. We fully test and evaluate on three widely used large-scale benchmark datasets, and the results show that our model outperforms existing COD models.

关键词： camouflaged object detection Feature aggregation Multi-scale features Global semantic fusion

来源：评论

学校读者我要写书评

暂无评论

camouflaged object detection Based on Localization Guidance and Multi-scale Refinement 31st

Camouflaged Object Detection Based on Localization Guidance ...

引用

31st International Conference on Multimedia Modeling

作者： Wang, Jinyang Wu, Wei Inner Mongolia Univ Hohhot Peoples R China

ISBN: (纸本)9789819620531;9789819620548

camouflaged object detection (COD) aims to visually segment camouflaged objects that blend into their surrounding environment. The problems of inaccurate object localization and lack of detailed information necessary for precise segmentation have not been effectively resolved in current camouflaged object detection models, and these models generally have a large number of parameters. To address these issues, we propose a localization guidance and multi-scale refinement network (LRNet). Through effective aggregation and use of contextual and gradient information, the LRNet can achieve accurate localization and refined segmentation of camouflaged objects. Specifically, we develop a regional localization decoder (RLD) to locate the camouflaged object region by fusing multi-stage contextual information. Additionally, we employ an attention complementary fusion (ACF) module to enhance internal feature representation, which leverages gradient features to compensate for the limitations of contextual features in geometric texture expression. To accurately distinguish the camouflaged object from the background, we propose an iterative refinement fusion (IRF) module. The IRF module integrates adjacent-level features under the guidance of localization information to refine details at multiple scales. Extensive experimental results demonstrate the efficacy of our LRNet, which employs only 22.33M parameters and significantly outperforms SOTA methods on three widely used benchmark datasets.

关键词： camouflaged object detection Contextual information Gradient feature object localization

来源：评论

学校读者我要写书评

暂无评论

camouflaged object detection via Scale-Feature Attention and Type-Feature Attention 7th

Camouflaged Object Detection via Scale-Feature Attention and...

引用

7th Chinese Conference on Pattern Recognition and Computer Vision

作者： Liu, Yi Meng, Hui Changzhou Univ Sch Comp Sci & Artificial Intelligence Aliyun Sch Big Data Changzhou Peoples R China Changzhou Univ Sch Software Changzhou Peoples R China

ISBN: (纸本)9789819788576;9789819788583

camouflaged object detection targets at identifying and segmenting objects hidden in the surroundings. Due to the various shapes and sizes, and highly non-discriminative features of camouflaged objects, it is a challenge for Convolutional Neural Networks (CNNs) to detect them from the background. To tackle the first problem of various shapes and sizes, we propose a Scale-Feature Attention (SFA), which can effectively integrate feature information of different scales, so that the model can comprehensively perceive and understand the visual characteristics of different sizes of camouflaged objects. Additionally, the traditional CNN model is difficult to capture the part-whole relationship of camouflaged objects. To solve the second problem of CNNs' limitation, we propose a Type-Feature Attention (TFA) to integrate contrast from CNNs and part-whole relations from CapsNets, which will improve the identification and object wholeness of camouflaged objects. Experiments on three camouflaged object detection benchmark datasets show that both the proposed SFA and TFA achieve significant performance improvement, which verifies the superiority of the proposed method.

关键词： camouflaged object detection Scale-feature attention Type-feature attention CapsNets

来源：评论

学校读者我要写书评

暂无评论

COMPrompter: reconceptualized segment anything model with multiprompt network for camouflaged object detection

引用

Science China(Information Sciences) 2025年第1期68卷 189-203页

作者： Xiaoqin ZHANG Zhenni YU Li ZHAO Deng-Ping FAN Guobao XIAO Zhejiang Province Key Laboratory of Intelligent Informatics for Safety and Emergency Wenzhou University Nankai International Advanced Research Institute (SHENZHEN FUTIAN) College of Computer Science Nankai University School of Computer Science and Technology Tongji University

We rethink the segment anything model(SAM) and propose a novel multiprompt network called COMPrompter for camouflaged object detection(COD). SAM has zero-shot generalization ability beyond other models and can provide an ideal framework for COD. Our network aims to enhance the single prompt strategy in SAM to a multiprompt strategy. To achieve this, we propose an edge gradient extraction module, which generates a mask containing gradient information regarding the boundaries of camouflaged objects. This gradient mask is then used as a novel boundary prompt, enhancing the segmentation process. Thereafter, we design a box-boundary mutual guidance module, which fosters more precise and comprehensive feature extraction via mutual guidance between a boundary prompt and a box prompt. This collaboration enhances the model's ability to accurately detect camouflaged objects. Moreover, we employ the discrete wavelet transform to extract high-frequency features from image embeddings. The high-frequency features serve as a supplementary component to the multiprompt ***, our COMPrompter guides the network to achieve enhanced segmentation results, thereby advancing the development of SAM in terms of COD. Experimental results across COD benchmarks demonstrate that COMPrompter achieves a cutting-edge performance, surpassing the current leading model by an average positive metric of 2.2% in COD10K. In the specific application of COD, the experimental results in polyp segmentation show that our model is superior to top-tier methods as well. The code will be made available at https://***/guobaoxiao/COMPrompter.

关键词： segment anything model camouflaged object detection boundary prompt

来源：评论

学校读者我要写书评

暂无评论

Dynamic adaptive scaling network for camouflaged object detection

引用

SIGNAL IMAGE AND VIDEO PROCESSING 2025年第1期19卷 1-9页

作者： Wang, Lili Ruan, Kang Yan, Pu Zhao, Yang Wang, Xu Anhui Jianzhu Univ Sch Elect & Informat Engn Hefei 230601 Anhui Peoples R China

Recently, accurately segmenting objects from backgrounds has become a significant challenge for camouflaged object detection. Currently, traditional single-view approaches have limitations in predicting the boundary of camouflaged objects. Moreover, a fixed scaling ratio fails to fully explore imperceptible clues between camouflaged objects and background surroundings. To overcome these obstacles, we propose a triplet network via dynamic adaptive scaling strategy, named Dynamic Adaptive Scaling Network for camouflaged object detection(DASNet). It mimics the behavior of humans dynamically adjusting observation distances when observing indistinct objects to zoom in and out images. Specifically, we design a dynamic adaptive scaling module to obtain the optimal scaling ratios for each view. In addition, we design the multi-view feature fusion module and the multi-scale feature fusion module to guide the learning of multi-view and multi-scale features and enhance feature representation. Extensive experiments demonstrate that the proposed DASNet outperforms 24 other representative methods on three public datasets.

关键词： camouflaged object detection Dynamic adaptive scaling strategy Multi-scale feature fusion module

来源：评论

学校读者我要写书评

暂无评论

Multi-task information propagation network for camouflaged object detection

引用

DIGITAL SIGNAL PROCESSING 2025年 162卷

作者： Zhao, Fei Lou, Wenzhong Feng, Hengzhen Ding, Nanxi Li, Chenglong Ma, Wenlong Beijing Inst Technol Sch Mechatron Engn Beijing 100081 Peoples R China

camouflaged object detection (COD) aims to discover camouflaged objects embedded in the background. However, most existing COD methods focus solely on extracting single-task features for inference, and rely on relatively simple supervision data, which hampers their generalization ability in real-world deployment. To meet these challenges, we design a novel camouflaged object detection framework, the Multi-Task Information Propagation Network (MIP-Net). In the feature extraction stage, the Multi-Task Feature Extraction Module (MFEM) learns multi-task feature representations that preserve boundary, texture, and object information to enhance COD performance. During the feature fusion encoding stage, the Cross-Level Multi-Task Fusion Propagation (CMFP) module effectively integrates multi-task features from different levels, adaptively weighting the encoder's features so that the decoder receives more effective information for improved COD outcomes. In the feature decoding inference stage, the Cross-Layer Progressive Refinement Module (CSRM) captures the complementarity of cross-level features, further refining the final inference results. We conducted comprehensive experiments on four benchmark datasets: CHAMELEON, CAMO, COD10K, and NC4K. The comprehensive results prove that the proposed MIP-Net framework outperforms existing methods, showcasing its effectiveness and robustness.

关键词： camouflaged object detection Swin transformer Multi-task learning Feature fusion Feature propagation

来源：评论

学校读者我要写书评

暂无评论

When CNN meet with ViT: decision-level feature fusion for camouflaged object detection

引用

VISUAL COMPUTER 2025年第6期41卷 3957-3972页

作者： Yue, Guowen Jiao, Ge Li, Chen Xiang, Jiahao Hengyang Normal Univ Coll Comp Sci & Technol Hengyang 421002 Peoples R China Hengyang Normal Univ Hunan Prov Key Lab Intelligent Informat Proc & App Hengyang 421002 Peoples R China

Despite the significant advancements in camouflaged object detection achieved by convolutional neural network (CNN) methods and vision transformer (ViT) methods, both have limitations. CNN-based methods fail to explore long-range dependencies due to their limited receptive fields, while ViT-based methods lose detailed information due to large-span aggregation. To address these issues, we introduce a novel model, the double-extraction and triple-fusion network (DTNet), which leverages the global context modeling capabilities of ViT-based encoders and the detail capture capabilities of CNN-based encoders through decision-level feature fusion to make up the respective shortcomings for more complete segmentation of camouflaged objects. Specifically, it incorporates a boundary guidance module, designed to aggregate high-level and low-level boundary information through multi-scale feature decoding, thereby guiding the local detail representation of the transformer. It also includes a global context aggregation module, which shrinks the information of adjacent channels from top to bottom and aggregates information of high-level and low-level scales from bottom to top for feature decoding. It also contains a multi-feature fusion module to fuse global context features and local detail features. This module employs the attention mechanism in different channels to assign varying weights to long-range and short-range information. Through extensive experimentation, it has proven that the DTNet significantly surpasses 20 recently state-of-the-art methods in terms of performance. The related code and datasets will be posted at https://***/KungFuProgrammerle/DTNet.

关键词： Convolutional neural network Vision transformer camouflaged object detection Feature fusion

来源：评论

学校读者我要写书评

暂无评论

IRFNet: Cognitive-Inspired Iterative Refinement Fusion Network for camouflaged object detection

引用

SENSORS 2025年第5期25卷 1555-1555页

作者： Li, Guohan Wang, Jingxin Wei, Jianming Xu, Zhengyi Chinese Acad Sci Shanghai Adv Res Inst Shanghai 201210 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China ShanghaiTech Univ Sch Informat Sci & Technol Shanghai 201210 Peoples R China

camouflaged object detection (COD) aims to identify objects that are intentionally concealed within their surroundings through appearance, texture, or pattern adaptations. Despite recent advances, extreme object-background similarity causes existing methods struggle with accurately capturing discriminative features and effectively modeling multiscale patterns while preserving fine details. To address these challenges, we propose Iterative Refinement Fusion Network (IRFNet), a novel framework that mimics human visual cognition through progressive feature enhancement and iterative optimization. Our approach incorporates the following: (1) a Hierarchical Feature Enhancement Module (HFEM) coupled with a dynamic channel-spatial attention mechanism, which enriches multiscale feature representations through bilateral and trilateral fusion pathways;and (2) a Context-guided Iterative Optimization Framework (CIOF) that combines transformer-based global context modeling with iterative refinement through dual-branch supervision. Extensive experiments on three challenging benchmark datasets (CAMO, COD10K, and NC4K) demonstrate that IRFNet consistently outperforms fourteen state-of-the-art methods, achieving improvements of 0.9-13.7% across key metrics. Comprehensive ablation studies validate the effectiveness of each proposed component and demonstrate how our iterative refinement strategy enables progressive improvement in detection accuracy.

关键词： object detection computer vision camouflaged object detection attention mechanism iterative refinement cross-level feature fusion

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：