Pattern Recognition最新文献_第4页

Prototype-augmented mean teacher for robust semi-supervised medical image segmentation 鲁棒半监督医学图像分割的原型增强均值教师

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-17 DOI: 10.1016/j.patcog.2025.111722

Huaikun Zhang , Pei Ma , Jizhao Liu , Jing Lian , Yide Ma

{"title":"Prototype-augmented mean teacher for robust semi-supervised medical image segmentation","authors":"Huaikun Zhang , Pei Ma , Jizhao Liu , Jing Lian , Yide Ma","doi":"10.1016/j.patcog.2025.111722","DOIUrl":"10.1016/j.patcog.2025.111722","url":null,"abstract":"<div><div>Semi-supervised learning has made significant progress in medical image segmentation, aiming to improve model performance with small amounts of labeled data and large amounts of unlabeled data. However, most existing methods focus too much on the supervision of label space and have insufficient supervision on feature space. Moreover, these methods generally focus on enhancing inter-class discrimination, ignoring the processing of intra-class variation, which has significant effects on fine-grained segmentation in complex medical images. To overcome these limitations, we propose a novel semi-supervised segmentation approach, Prototype-Augmented Mean Teacher (PAMT). Built upon the Mean Teacher framework, PAMT incorporates non-learnable prototypes to enhance feature space supervision. Specifically, we introduce two innovative loss functions: Prototype-Guided Pixel Classification (PGPC) Loss and Adaptive Prototype Contrastive (APC) Loss. PGPC Loss ensures pixel classification consistency with the nearest prototypes through a nearest-neighbor strategy, while APC Loss further captures intra-class variability, thereby improving the model's capacity to distinguish between pixels of the same class. By augmenting the Mean Teacher framework with prototype learning, PAMT not only improves feature representation and mitigates pseudo-label noise but also boosts segmentation accuracy and generalization, particularly in complex anatomical structures. Extensive experiments on three public datasets demonstrate that PAMT consistently surpasses state-of-the-art methods.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111722"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

BE-ECM: Belief Entropy-based Evidential C-Means and its application in data clustering 基于信念熵的证据c均值及其在数据聚类中的应用

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-17 DOI: 10.1016/j.patcog.2025.111676

Jixiang Deng , Guohui Zhou , Yong Deng , Kang Hao Cheong

{"title":"BE-ECM: Belief Entropy-based Evidential C-Means and its application in data clustering","authors":"Jixiang Deng , Guohui Zhou , Yong Deng , Kang Hao Cheong","doi":"10.1016/j.patcog.2025.111676","DOIUrl":"10.1016/j.patcog.2025.111676","url":null,"abstract":"<div><div>As an extension of Fuzzy C-Means based on Dempster-Shafer evidence theory, Evidential C-Means (ECM) generalizes fuzzy partition to credal partition and has been widely applied. However, ECM’s objective function only considers distortion between objects and prototypes, making it highly sensitive to prototype initialization and prone to the local optima problem. While maximum entropy-based methods improve stability by entropy regularization, they are limited to fuzzy partition and cannot handle credal partition with multi-class uncertainty in evidential clustering. To overcome the issues, this paper proposes Belief Entropy-based Evidential C-Means (BE-ECM), which uniquely equips ECM with a belief entropy-based Maximum Entropy Principle (MEP) framework. Compared to ECM, BE-ECM considers not only the distortion term but also a negative belief entropy term, leveraging MEP to enhance stability against the local optimal problem. Unlike other maximum entropy-based methods, BE-ECM incorporates credal partition with belief entropy, enabling explicit multi-class uncertainty modeling and stable evidential clustering. During the clustering process of BE-ECM, the negative belief entropy term initially dominates to provide unbiased estimation for unknown data distributions, mitigating the impact of poorly initialized prototypes and reducing the risks of local optima, while the distortion term gradually refines the credal partition as clustering progresses. Experimental results demonstrate BE-ECM’s superior performance and high stability on clustering tasks compared with the existing clustering algorithms.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"167 ","pages":"Article 111676"},"PeriodicalIF":7.5,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143876928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Class and Domain Low-rank Tensor Learning for Multi-source Domain Adaptation 多源域自适应的类和域低秩张量学习

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-16 DOI: 10.1016/j.patcog.2025.111675

Yuwu Lu , Huiling Fu , Zhihui Lai , Xuelong Li

{"title":"Class and Domain Low-rank Tensor Learning for Multi-source Domain Adaptation","authors":"Yuwu Lu , Huiling Fu , Zhihui Lai , Xuelong Li","doi":"10.1016/j.patcog.2025.111675","DOIUrl":"10.1016/j.patcog.2025.111675","url":null,"abstract":"<div><div>Multi-source unsupervised domain adaptation (MUDA) aims to transfer knowledge from multiple labeled source domains to an unlabeled target domain. A key challenge in MUDA is to minimize the distributional discrepancy between the source and target domains. While traditional methods typically merge source domains to reduce this discrepancy, they often overlook higher-order correlations and class-discriminative relationships across domains, which weakens the generalization and classification abilities of the model. To address these challenges, we propose a novel method called Class and Domain Low-rank Tensor Learning (CDLTL), which integrates domain-level alignment and class-level alignment into a unified framework. Specifically, CDLTL leverages a projection matrix to map data from both source and target domains into a shared subspace, enabling the reconstruction of target domain samples from the source data and thereby reducing domain discrepancies. By combining tensor learning with joint sparse and weighted low-rank constraints, CDLTL achieves domain-level alignment, allowing the model to capture complex higher-order correlations across multiple domains while preserving global structures within the data. CDLTL also takes into account the geometric structure of multiple source domains and preserves local structures through manifold learning. Additionally, CDLTL achieves class-level alignment through class-based low-rank constraints, which improve intra-class compactness and inter-class separability, thus boosting the discriminative ability and robustness of the model. Extensive experiments conducted across various visual domain adaptation tasks demonstrate that the proposed method outperforms some of the existing approaches.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"167 ","pages":"Article 111675"},"PeriodicalIF":7.5,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143878533","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DecloudFormer: Quest the key to consistent thin cloud removal of wide-swath multi-spectral images DecloudFormer：寻找宽幅多光谱图像中一致的薄云去除的关键

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-15 DOI: 10.1016/j.patcog.2025.111664

Mingkai Li , Qizhi Xu , Kaiqi Li , Wei Li

{"title":"DecloudFormer: Quest the key to consistent thin cloud removal of wide-swath multi-spectral images","authors":"Mingkai Li , Qizhi Xu , Kaiqi Li , Wei Li","doi":"10.1016/j.patcog.2025.111664","DOIUrl":"10.1016/j.patcog.2025.111664","url":null,"abstract":"<div><div>Wide-swath images contain clouds of various shapes and thicknesses. Existing methods have different thin cloud removal strengths in different patches of the wide-swath image. This leads to severe cross-patch color inconsistency in the thin cloud removal results of wide-swath images. To solve this problem, a DecloudFormer with cross-patch thin cloud removal consistency was proposed. First, a Group Layer Normalization (GLNorm) was proposed to preserve both the spatial and channel distribution of thin cloud. Second, a CheckerBoard Mask (CB Mask) was proposed to make the network focus on different cloud-covered areas of the image and extract local cloud features. Finally, a two-branch DecloudFormer Block containing the CheckerBoard Attention (CBA) was proposed to fuse the global cloud features and local cloud features to reduce the cross-patch color difference. DecloudFormer and compared methods were tested for simulated thin cloud removal performance on images from QuickBird, GaoFen-2, and WorldView-2 satellites, and for real thin cloud removal performance on images from Landsat-8 satellite. The experiment results demonstrated that DecloudFormer outperformed the existing State-Of-The-Art (SOTA) methods. Furthermore, DecloudFormer makes it possible to process thin cloud covered wide-swath image using a small video memory GPU. The source code are available at <span><span>the link</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111664"},"PeriodicalIF":7.5,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Domain adaptive depth completion via spatial-error consistency 基于空间误差一致性的域自适应深度补全

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-15 DOI: 10.1016/j.patcog.2025.111645

Lingyu Xiao , Jinhui Wu , Junjie Hu , Ziyu Li , Wankou Yang

{"title":"Domain adaptive depth completion via spatial-error consistency","authors":"Lingyu Xiao , Jinhui Wu , Junjie Hu , Ziyu Li , Wankou Yang","doi":"10.1016/j.patcog.2025.111645","DOIUrl":"10.1016/j.patcog.2025.111645","url":null,"abstract":"<div><div>In this paper, we introduce a novel training framework designed to address the challenge of unsupervised domain adaptation (UDA) in depth completion. Our framework aims to bridge the gap between lidar and image data by establishing a shared domain, which is a collection of the confidence of the network’s prediction. By indirectly adapting the depth network through this common domain, the problem is decomposed into two key tasks: (1) constructing the common domain and (2) adapting the depth network using the common domain. For the construction of the common domain, errors in the network’s predictions are modelled as confidence, which serves as supervision for a sub-module called the Depth Completion Plugin (DCPlugin). The purpose of the DCPlugin is to generate the confidence associated with any given dense depth prediction. To adapt the depth network using the common domain, a confidence-aware co-training task is employed, leveraging the confidence map provided by the well-adapted DCPlugin. To assess the effectiveness of our proposed approach, we conduct experiments on multiple depth networks under adaptation scenarios, namely CARLA <span><math><mo>→</mo></math></span> KITTI and VKITTI <span><math><mo>→</mo></math></span> KITTI. The results demonstrate that our method surpasses other domain adaptation (DA) techniques, achieving state-of-the-art performance. Given the limited existing work in this domain, we provide comprehensive discussions to guide future researchers in this field.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111645"},"PeriodicalIF":7.5,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143839340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Rethinking transformers with convolution and graph embeddings for few-shot molecular property discovery 基于卷积和图嵌入的小波分子性质发现的再思考

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-15 DOI: 10.1016/j.patcog.2025.111657

Luis H.M. Torres, Joel P. Arrais, Bernardete Ribeiro

{"title":"Rethinking transformers with convolution and graph embeddings for few-shot molecular property discovery","authors":"Luis H.M. Torres, Joel P. Arrais, Bernardete Ribeiro","doi":"10.1016/j.patcog.2025.111657","DOIUrl":"10.1016/j.patcog.2025.111657","url":null,"abstract":"<div><div>The prediction of molecular properties is a critical step in drug discovery campaigns. Computational methods such as graph neural networks (GNNs) and Transformers have effectively leveraged the small-range and long-range dependencies in molecules to preserve the local and global patterns for multiple molecular property prediction tasks. However, the dependence of these models on large amounts of experimental data poses a challenge, particularly on smaller biological datasets prevalent across the drug discovery pipeline. This paper introduces FS-GCvTR, a few-shot graph-based convolutional Transformer architecture designed to predict chemical properties with a small amount of labeled compounds. The convolutional Transformer is presented as a crucial component, effectively integrating both local and global dependencies of molecular graph embeddings by propagating a set of convolutional tokens across Transformer attention layers for molecular property prediction. Furthermore, a few-shot meta-learning approach is introduced to iteratively adapt model parameters across multiple few-shot tasks while generalizing to new chemical properties with limited available data. Experiments including few-shot evaluations on multi-property datasets show that the FS-GCvTR model outperformed other few-shot graph-based baselines in specific molecular property prediction tasks.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111657"},"PeriodicalIF":7.5,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143839328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IFShip: Interpretable fine-grained ship classification with domain knowledge-enhanced vision-language models IFShip：基于领域知识增强的视觉语言模型的可解释细粒度船舶分类

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-15 DOI: 10.1016/j.patcog.2025.111672

Mingning Guo, Mengwei Wu, Yuxiang Shen, Haifeng Li, Chao Tao

{"title":"IFShip: Interpretable fine-grained ship classification with domain knowledge-enhanced vision-language models","authors":"Mingning Guo, Mengwei Wu, Yuxiang Shen, Haifeng Li, Chao Tao","doi":"10.1016/j.patcog.2025.111672","DOIUrl":"10.1016/j.patcog.2025.111672","url":null,"abstract":"<div><div>End-to-end interpretation currently dominates the remote sensing fine-grained ship classification (RS-FGSC) task. However, the inference process remains uninterpretable, leading to criticisms of these models as “black box” systems. To address this issue, we propose a domain knowledge-enhanced Chain-of-Thought (CoT) prompt generation mechanism, which is used to semi-automatically construct a task-specific instruction-following dataset, TITANIC-FGS. By training on TITANIC-FGS, we adapt general-domain vision-language models (VLMs) to the FGSC task, resulting in a model named IFShip. Building upon IFShip, we develop an FGSC visual chatbot that redefines the FGSC problem as a step-by-step reasoning task and conveys the reasoning process in natural language. Experimental results show that IFShip outperforms state-of-the-art FGSC algorithms in both interpretability and classification accuracy. Furthermore, compared to VLMs such as LLaVA and MiniGPT-4, IFShip demonstrates superior performance on the FGSC task. It provides an accurate chain of reasoning when fine-grained ship types are recognizable to the human eye and offers interpretable explanations when they are not.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111672"},"PeriodicalIF":7.5,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143839384","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A dynamic predictive transformer with temporal relevance regression for action detection 动态预测变压器与时间相关回归的动作检测

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-14 DOI: 10.1016/j.patcog.2025.111644

Matthew Korban , Peter Youngs , Scott T. Acton

{"title":"A dynamic predictive transformer with temporal relevance regression for action detection","authors":"Matthew Korban , Peter Youngs , Scott T. Acton","doi":"10.1016/j.patcog.2025.111644","DOIUrl":"10.1016/j.patcog.2025.111644","url":null,"abstract":"<div><div>This paper introduces a novel transformer network tailored to skeleton-based action detection in untrimmed long video streams. Our approach centers around three innovative mechanisms that collectively enhance the network’s temporal analysis capabilities. First, a new predictive attention mechanism incorporates future frame data into the sequence analysis during the training phase. This mechanism addresses the essential issue of the current action detection models: incomplete temporal modeling in long action sequences, particularly for boundary frames that lie outside the network’s immediate temporal receptive field, while maintaining computational efficiency. Second, we integrate a new adaptive weighted temporal attention system that dynamically evaluates the importance of each frame within an action sequence. In contrast to the existing approaches, the proposed weighting strategy is both adaptive and interpretable, making it highly effective in handling long sequences with numerous non-informative frames. Third, the network incorporates an advanced regression technique. This approach independently identifies the start and end frames based on their relevance to different frames. Unlike existing homogeneous regression methods, the proposed regression method is heterogeneous and based on various temporal relationships, including those in future frames in actions, making it more effective for action detection. Extensive experiments on prominent untrimmed skeleton-based action datasets, PKU-MMD, OAD, and the Charade dataset demonstrate the effectiveness of this network.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111644"},"PeriodicalIF":7.5,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Secure reversible privacy protection for face multiple attribute editing 安全可逆的隐私保护的脸多属性编辑

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-12 DOI: 10.1016/j.patcog.2025.111662

Yating Zeng, Xinpeng Zhang, Guorui Feng

{"title":"Secure reversible privacy protection for face multiple attribute editing","authors":"Yating Zeng, Xinpeng Zhang, Guorui Feng","doi":"10.1016/j.patcog.2025.111662","DOIUrl":"10.1016/j.patcog.2025.111662","url":null,"abstract":"<div><div>The demand for face attribute editing is increasing across various applications, such as digital media and virtual reality. However, while existing methods can achieve high-quality multi-attribute editing, they often struggle to balance privacy protection and image reversibility, and are prone to causing undesired changes in non-target attributes. To address these issues, we propose a novel Multi-Layer Mapping and Password Fusion (M-LMPF) framework for efficient and flexible face attribute editing with reversible privacy protection. Our approach integrates multi-attribute editing with secure reversible image attribute protection, enabling precise control over the modification of target attributes while preserving facial identity consistency and avoiding changes to other attributes. The framework employs a deep multi-layer latent mapping network that embeds password information at different granular levels in the latent space, allowing for fine-grained control over facial features. Additionally, we introduce a new encryption and decryption mechanism to ensure reversible editing of specific attributes, effectively preventing unauthorized access. Extensive experiments demonstrate that the M-LMPF framework outperforms state-of-the-art methods in attribute editing accuracy, reversibility, identity consistency, and image quality.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111662"},"PeriodicalIF":7.5,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143855128","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Separation of Unknown Features and Samples for Unbiased Source-free Open Set Domain Adaptation 基于无偏无源开放集域自适应的未知特征与样本分离

IF 7.5 1区计算机科学

Pattern Recognition Pub Date : 2025-04-12 DOI: 10.1016/j.patcog.2025.111661

Fu Li , Yifan Lan , Yuwu Lu , Wai Keung Wong , Ming Zhao , Zhihui Lai , Xuelong Li

{"title":"Separation of Unknown Features and Samples for Unbiased Source-free Open Set Domain Adaptation","authors":"Fu Li , Yifan Lan , Yuwu Lu , Wai Keung Wong , Ming Zhao , Zhihui Lai , Xuelong Li","doi":"10.1016/j.patcog.2025.111661","DOIUrl":"10.1016/j.patcog.2025.111661","url":null,"abstract":"<div><div>Open Set Domain Adaptation (OSDA) is proposed to train a model on a source domain that performs well on a target domain with domain discrepancy and unknown class samples outside the source domain. Recently, Source-free Open Set Domain Adaptation (SF-OSDA) aims to achieve OSDA without accessing source domain samples. Existing SF-OSDA only focuses on the known class samples in the target domain and overlooks the abundant unknown class semantics in the target domain. To address these issues, in this paper, we propose a Separation of Unknown Features and Samples (SUFS) method for unbiased SF-OSDA. Specifically, SUFS consists of a Sample Feature Separation (SFS) module that separates the private features from the known features in each sample. This module not only utilizes the semantic information of each sample label, but also explores the potential unknown information of each sample. Then, we integrate a Feature Correlation Representation (FCR) module, which computes the similarity between each sample and its neighboring samples to correct semantic bias and build instance-level decision boundaries. A large number of experiments in the SF-OSDA scenario have demonstrated the effectiveness of SUFS. In addition, SUFS also shows great performance in the Source-free Partial Domain Adaptation (SF-PDA) scenario.</div></div>","PeriodicalId":49713,"journal":{"name":"Pattern Recognition","volume":"166 ","pages":"Article 111661"},"PeriodicalIF":7.5,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143843011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0