Information Fusion最新文献

{"title":"A two-stage learning network for PVINS modeling and fusion estimation in challenging environments","authors":"Xuanyu Wu ,&nbsp;Jiankai Yin ,&nbsp;Jian Yang ,&nbsp;Xin Liu ,&nbsp;Wenshuo Li ,&nbsp;Lei Guo","doi":"10.1016/j.inffus.2026.104192","DOIUrl":"10.1016/j.inffus.2026.104192","url":null,"abstract":"<div><div>In the polarization-based visual-inertial navigation system (PVINS), information from polarization sensor (PS) and visual-inertial navigation system (VINS) is fused to enable position and attitude estimation, thereby offering an effective solution for autonomous navigation in global navigation satellite system (GNSS)-denied environments. However, under challenging conditions such as complex weather, the state-space model of PVINS becomes susceptible to uncertain model error, limiting the accuracy and adaptability of the system. To address this issue, we propose a tightly coupled PVINS integration scheme based on a two-stage learning network, which consists of model error compensation and adaptive Kalman gain learning. In the first stage, a deep neural network with a shared-weight architecture is designed to learn and compensate for the state-space model error, thereby reducing network complexity and enabling more precise system modeling. In the second stage, to improve fusion accuracy of PVINS, a Kalman gain learning network (KGLN)-based intelligent fusion method is proposed. This approach enables the adaptive learning of Kalman gains, circumventing the dependency of the system on knowledge of the noise statistics. Finally, the performance of the system is verified through the semi-physical simulation and flight test. The experimental results confirm that the proposed method outperforms conventional PVINS in terms of both position and heading estimation.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104192"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146089493","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}

{"title":"Validity-aware context modeling for gradient-guided image inpainting","authors":"Wuzhen Shi ,&nbsp;Wu Yang ,&nbsp;Zhihao Wu ,&nbsp;Yang Wen","doi":"10.1016/j.inffus.2026.104162","DOIUrl":"10.1016/j.inffus.2026.104162","url":null,"abstract":"<div><div>Existing prior-guided image inpainting methods show state-of-the-art performance. But their prior extraction is computationally expensive and unstable in accuracy. Besides, most of them only focus on the structure guidance, which hardly facilitates the repair of realistic textures. Inspired by the fact that gradient maps are easy to extract and reflect both image structure and fine texture details, this paper proposes a gradient-guided network for image inpainting, which first uses the gradient context information and multi-level image compensation features to repair the gradient, and then uses the repaired gradient features to guide the generation of realistic image. A gradient-driven attention (GDA) module is introduced for efficient prior guidance. Additionally, a context validity-aware (CVA) module is proposed for progressively filling hole regions of images, which accurately utilizes both local and contextual information for image inpainting via validity-aware measurements. Furthermore, by artificially manipulating the generation of the gradient map, our gradient-guided image inpainting method enables user-guided image editing, which effectively increases the diversity of image generation and enhances the flexibility of image editing. Experiments on benchmark datasets show that the proposed method outperforms the state-of-the-art methods. Extensive ablation experiments are also conducted to demonstrate the effectiveness of each module.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104162"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146000897","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}

{"title":"A novel knowledge distillation method for graph neural networks with gradient mapping and fusion","authors":"Kang Liu ,&nbsp;Shunzhi Yang ,&nbsp;Chang-Dong Wang ,&nbsp;Yunwen Chen ,&nbsp;Zhenhua Huang","doi":"10.1016/j.inffus.2026.104163","DOIUrl":"10.1016/j.inffus.2026.104163","url":null,"abstract":"<div><div>The primary goal of graph knowledge distillation (GKD) is to transfer knowledge from a complex graph neural network (GNN) teacher to a smaller, yet more efficient GNN or multi-layer perceptron student. Although existing methods address network scalability, they rely on a frozen teacher that fails to explain how to derive results, thus limiting performance and hindering the improvement of a student. Therefore, we propose a novel GKD method, termed Dynamic Gradient Distillation (DGD), consisting of Generative Adversarial Imitation Learning (GAIL)-based Gradient Mapping and Two-Stage Gradient Fusion modules. The former builds the teacher’s learning process to understand knowledge by drawing on the principle of GAIL. The latter consists of attention fusion and weighted bias operations. Through the attentional fusion operation, it captures and fuses the responses of the teacher to change the gradient of the student at each layer. The fused gradients are then updated by combining them with the student’s backpropagated gradients using the weighted bias operation. DGD allows the student to inherit and extend the teacher’s learning process efficiently. Extensive experiments conducted with seven publicly available datasets show that DGD could significantly outperform some existing methods in node classification tasks. Our code and data are released at <span><span>https://github.com/KangL-G/Dynamic-Gradient-Distillation</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104163"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146006526","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}

{"title":"Arbitrary-scale spatial-spectral fusion using kernel integral and progressive resampling","authors":"Wei Li,&nbsp;Honghui Xu,&nbsp;Yueqian Quan,&nbsp;Zhe Chen,&nbsp;Jianwei Zheng","doi":"10.1016/j.inffus.2026.104143","DOIUrl":"10.1016/j.inffus.2026.104143","url":null,"abstract":"<div><div>Benefiting from the booming deep learning techniques, spatial-spectral fusion (SSF) is considered as an ideal alternative to break the traditions of acquiring hyperspectral images (HSI) with costly devices. Yet with the remarkable progress, current solutions necessitate training and storing multiple models for different scaling factors. To overcome this dilemma, we propose a spatial-spectral fusion neural operator (SFNO) to perform <strong>arbitrary-scale</strong> SSF within the operator learning framework. Specifically, SFNO approaches the problem from the perspective of approximation theory by embedding the features of two degraded functions into a high-dimensional latent space through pointwise convolution layers, thereby capturing richer spectral feature information. Consequently, the mapping between function spaces is approximated via the Galerkin integral (GI) mechanism, which culminates in a final dimensionality reduction step to produce a high-resolution HSI. Moreover, we propose a progressive resampling integration (PR) that resamples the integrand’s domain in the triple kernel integration to provide non-local multi-scale information. The synergistic action of both integration mechanisms enables SFNO to effortlessly handle magnification factors it never encountered during training. Extensive experiments on the CAVE, Chikusei, Pavia Centre, Harvard, and real-world datasets demonstrate that our SFNO delivers substantial improvements over existing state-of-the-art methods. In particular, under the 8× upsampling setting on the CAVE, Chikusei, and Pavia Centre datasets, SFNO surpasses the second-best model by 0.56 dB, 1.05 dB, and 0.72 dB in PSNR, respectively. Our code is publicly available at <span><span>https://github.com/weili419/SFNO</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104143"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145957301","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}

{"title":"MCIVA: A multi-view pedestrian detection framework with a central inverse nearest neighbor map and a view adaptive module","authors":"He Li ,&nbsp;Taiyu Liao ,&nbsp;Weihang Kong ,&nbsp;Xingchen Zhang","doi":"10.1016/j.inffus.2026.104142","DOIUrl":"10.1016/j.inffus.2026.104142","url":null,"abstract":"<div><div>Multi-view pedestrian detection is an important task and has many applications in areas such as surveillance and smart cities. Despite the significant performance improvements achieved in recent multi-view pedestrian detection methods, there are still three main challenges for this task. First, in crowded areas, neighboring connected components may merge in dense regions, resulting in unclear localization of pixel peaks for each pedestrian. Second, the loss functions used in previous multi-view pedestrian detection methods have a high response to background regions. Third, camera parameters have not been fully utilized; they are only used to generate fixed projection matrices. To address these challenges, we propose a novel multi-view pedestrian detection framework (<strong>MCIVA</strong>) with a central inverse nearest neighbor <em>(CINN)</em> map and a view adaptive module <em>(VAM)</em>. The CINN map is introduced to generate the ground-truth probability occupancy map (POM) based on annotations, providing more precise location information for each pedestrian. To enhance the model’s attention to local structural information, we propose <em>a local structural similarity loss</em> to reduce the influence of false local maxima in background regions. Moreover, the VAM is introduced to utilize camera parameters to generate learnable weights for multi-view feature fusion. We evaluate the proposed method on three benchmark datasets, and the results show that the proposed MCIVA improves the quality of prediction maps and achieves state-of-the-art performance.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104142"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145957305","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}

{"title":"A data fusion approach to synthesize microwave imagery of tropical cyclones from infrared data using vision transformers","authors":"Fan Meng ,&nbsp;Tao Song ,&nbsp;Xianxuan Lin ,&nbsp;Kunlin Yang","doi":"10.1016/j.inffus.2026.104167","DOIUrl":"10.1016/j.inffus.2026.104167","url":null,"abstract":"<div><div>Microwave images with high spatiotemporal resolution are essential for observing and predicting tropical cyclones (TCs), including TC positioning, intensity estimation, and detection of concentric eyewall. Nevertheless, the temporal resolution of tropical cyclone microwave (TCMW) images is limited due to satellite quantity and orbit constraints, presenting a challenging problem for TC disaster forecasting. This research suggests a multi-sensor data fusion approach, using high-temporal-resolution tropical cyclone infrared (TCIR) images to generate synthetic TCMW images, offering a solution to this data scarcity problem. In particular, we introduce a deep learning network based on the Vision Transformer (TCA-ViT) to translate TCIR images into TCMW images. This can be viewed as a form of synthetic data generation, enhancing the available information for decision-making. We integrate a phase-based physical guidance mechanism into the training process. Furthermore, we have developed a dataset of TC infrared-to-microwave image conversions (TCIR2MW) for training and testing the model. Experimental results demonstrate the method’s capability in rapidly and accurately extracting key features of TCs. Leveraging techniques like Mask and Transfer Learning, it addresses the absence of TCMW images by generating MW images from IR images, thereby aiding downstream tasks like TC intensity and precipitation forecasting. This study introduces a novel approach to the field of TC image research, with the potential to advance deep learning in this direction and provide vital insights for real-time observation and prediction of global TCs. Our source code and data are publicly available online at <span><span>https://github.com/kleenY/TCIR2MW</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104167"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146033286","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}

{"title":"IDFL: Incentive-driven federated learning with selfish clients","authors":"Jin Xu,&nbsp;Hengrun Zhang,&nbsp;Huiqun Yu,&nbsp;Guisheng Fan","doi":"10.1016/j.inffus.2026.104185","DOIUrl":"10.1016/j.inffus.2026.104185","url":null,"abstract":"<div><div>Heterogeneity challenges have been long discussed in Federated Learning (FL). Among these challenges, statistical heterogeneity, where non-independent and identical (non-IID) data distributions across clients severely impact model convergence and performance, remains particularly problematic. While existing batch size optimization strategies effectively address system-level heterogeneity and resource constraints, they inadequately tackle statistical heterogeneity, often simply increasing batch sizes without theoretical justification. Such approaches overlook a critical convergence-generalization dilemma well-established in traditional machine learning: larger batch sizes accelerate convergence but may deteriorate generalization performance beyond critical thresholds, which is usually termed “generalization gap”. To bridge this gap in FL, we propose a comprehensive framework with three key contributions. First, we establish a batch size optimization mechanism that balances convergence and generalization objectives through a penalty function, providing mathematically derived closed-form solutions for optimal batch sizes. Second, we design a Stackelberg game-based incentive mechanism that coordinates batch size assignments with resource contributions while ensuring fair reward allocation to maximize individual client utility (defined as the difference between rewards and costs). Third, we develop a two-step verification strategy that detects and mitigates free-riding behaviors while monitoring convergence patterns to terminate ineffective training processes. Extensive experiments on real-world datasets validate our approach, demonstrating significant improvements in both convergence performance and fairness compared to state-of-the-art algorithms. Ablation studies confirm the effectiveness of each component.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104185"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146047983","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}

{"title":"Graph-guided cross-image correlation learning with adaptive global-local feature fusion for fine-grained visual representation","authors":"Hongxing You ,&nbsp;Yangtao Wang ,&nbsp;Xiaocui Li ,&nbsp;Yanzhao Xie ,&nbsp;Da Chen ,&nbsp;Xinyu Zhang ,&nbsp;Wensheng Zhang","doi":"10.1016/j.inffus.2026.104204","DOIUrl":"10.1016/j.inffus.2026.104204","url":null,"abstract":"<div><div>Fine-grained visual classification (FGVC) has been challenging due to the difficulty of distinguishing between highly similar local regions. Recent studies leverage graph neural network (GNN) to learn local representations, but they solely focus on patch interactions within each image, failing to capture semantic relationships across different samples and rendering fine-grained features semantically disconnected from each other. To address these challenges, we propose <strong>G</strong>raph-guided <strong>C</strong>ross-image <strong>C</strong>orrelation Learning with Adaptive Global-local Feature Fusion for Fine-grained Visual <strong>R</strong>epresentation (termed as GCCR). We design a Cross-image Correlation Learning (CCL) module where spatially corresponding patches across images are connected as graph nodes, enabling inter-image interactions to capture semantically rich local features. In this CCL module, we introduce a Ranking Loss to address the limitation of traditional classification losses that focus solely on maximizing individual sample confidence without explicitly constraining feature discriminability among visually similar categories. In addition, GCCR constructs a lightweight fusion module that dynamically balances the contributions of global and local features, leading to unbiased image representations. We conduct extensive experiments on 4 popular FGVC datasets including CUB-200-2011, Stanford Cars, FGVC-Aircraft, and iNaturalist 2017. Experimental results verify that GCCR can achieve much higher performance than the state-of-the-art (SOTA) FGVC methods, while maintaining lower model complexity. Take the most challenging iNaturalist 2017 for example, GCCR gains at least 7.51% accuracy while reducing more than 4.42M parameter scale and 80M FLOPs than the optimal solution. We release the pretrained model and code at GitHub: <span><span>https://github.com/dislie/GCCR</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104204"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146110180","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}

{"title":"Stain-aware domain alignment for imbalance blood cell classification","authors":"Yongcheng Li ,&nbsp;Lingcong Cai ,&nbsp;Ying Lu ,&nbsp;Xiao Han ,&nbsp;Ma Li ,&nbsp;Wenxing Lai ,&nbsp;Xiangzhong Zhang ,&nbsp;Xiaomao Fan","doi":"10.1016/j.inffus.2026.104166","DOIUrl":"10.1016/j.inffus.2026.104166","url":null,"abstract":"<div><div>Blood cell identification is critical for hematological analysis as it aids physicians in diagnosing various blood-related diseases. In real-world scenarios, blood cell image datasets often present the issues of domain shift and data imbalance, posing challenges for accurate blood cell identification. To address these issues, we propose a novel blood cell classification method termed SADA via stain-aware domain alignment. The primary objective of this work is to mine domain-invariant features in the presence of domain shifts and data imbalance. To accomplish this objective, we propose a stain-based augmentation approach and a local alignment constraint to learn domain-invariant features. Furthermore, we propose a domain-invariant supervised contrastive learning strategy to capture discriminative features. We decouple the training process into two stages of domain-invariant feature learning and classification training, alleviating the problem of data imbalance. Experiment results on four public blood cell datasets and a private real dataset collected from the Third Affiliated Hospital of Sun Yat-sen University demonstrate that SADA can achieve a new state-of-the-art baseline, which is superior to the existing cutting-edge methods. The source code can be available at the URL (<span><span>https://github.com/AnoK3111/SADA</span><svg><path></path></svg></span>).</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104166"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014547","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}

{"title":"Multimodal fusion of 3D point cloud and intraoperative imaging to enhance surgical robot navigation","authors":"Yiheng Wang ,&nbsp;Tianlun Wang ,&nbsp;Tao Liu ,&nbsp;Yihao Huang","doi":"10.1016/j.inffus.2026.104171","DOIUrl":"10.1016/j.inffus.2026.104171","url":null,"abstract":"<div><div>To address the insufficient navigation accuracy of surgical robots in dynamic, complex, and non-rigid intraoperative environments, this paper proposes an enhanced multimodal fusion framework—EMF-RSN. This framework achieves spatial consistency between point clouds and intraoperative images through a depth-guided Geometry-Vision Alignment Module (GVAN), implements dynamic weighted fusion of geometric and visual features through a cross-modal attention fusion module (CAFM), and constructs a closed-loop optimization mechanism from perception to decision through a Task Feedback Optimization (TFO) module, thereby improving navigation accuracy and stability.</div><div>Experiments on the public dataset (Hamlyn) and the self-built simulation dataset (Sim-Surgical Fusion) demonstrate that EMF-RSN significantly outperforms existing methods in terms of geometric accuracy, semantic consistency, and task robustness. Compared to traditional registration algorithms, point cloud errors are reduced by approximately 50%, trajectory errors are reduced by over 20%, and real-time performance of 44 FPS is maintained even in complex deformation and occlusion environments. This research provides a new technical approach and model foundation for realizing intelligent surgical navigation that integrates virtual and real elements, and is of great significance for the perception and autonomous control of surgical robots.</div></div>","PeriodicalId":50367,"journal":{"name":"Information Fusion","volume":"131 ","pages":"Article 104171"},"PeriodicalIF":15.5,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146014542","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}