Search Results (31)

In few-shot action recognition (FSAR), limited annotated data and large scene variations make it difficult for models to learn stable spatial semantics and reliable temporal dynamics. As a result, spatiotemporal representations tend to be weak, and models often fail to focus on discriminative motion regions or capture frame-to-frame changes accurately. Furthermore, the insufficient fusion of local details and global context renders the learned features more susceptible to background noise and scene bias. These issues become more pronounced when background clutter is severe or when different action classes share locally similar segments, leading to unreliable support–query matching and shifted similarity distributions, which ultimately result in class confusion. To address these challenges, we propose a bidirectional adaptive spatiotemporal modeling method integrated with contrastive learning for FSAR. The method constructs attention-guided bidirectional differencing features to model inter-frame variations with semantic alignment, while suppressing background noise. It introduces a local–global interactive channel attention module to strengthen both local and global dynamic representations, and integrates dynamic distance adjustment with hard negative mining during tuple-level matching. This combination imposes contrastive constraints that enhance intra-class compactness and inter-class separability, thereby mitigating interference from cross-class similar segments. Experiments under the standard 5-way 1-shot/5-shot protocol demonstrate consistent improvements across multiple datasets, and the proposed method achieves the best performance under the 5-shot setting while remaining competitive under the 1-shot setting. Full article

Few-shot open-set object detection (FS-OSOD) remains challenging in real-world scenarios, where detectors must accurately recognize known objects from few examples while reliably rejecting vast unknown categories. Under this setting, decision boundaries between known and unknown classes are easily distorted by data scarcity and background clutter, leading to severe overfitting on base classes and overconfident misclassification of unknowns. Recent research attempts to alleviate these issues by regularizing detection heads to suppress base-class bias, or by leveraging vision–language priors through open-vocabulary alignment and prompt tuning to enhance semantic transferability. However, these solutions often overlook explicit modeling of truly out-of-set unknowns and the instability of prompt adaptation in low-data regimes, which can cause boundary drifts and make unknown proposals be absorbed by similar seen classes or even suppressed as background. To alleviate these issues, a guided prompt–monument network (GPMN) that is proposed, which jointly enhances prompt learning and feature representation learning for FS-OSOD. First, the contrastive distilled prompts (CDP) module employs a teacher–student prompt framework to decouple optimization across base, novel, and unknown classes. This strategy preserves transferability between zero-shot and few-shot settings while enhancing discrimination on base categories. Second, a synthesized monument module (SMM) maintains class-centered memory with momentum-updated prototypes and a non-parametric classifier, which compresses the overlap between seen and unseen distributions and provides a stable rejection margin for unknowns with strong co-occurrence and background noise. Compared with existing head-regularization and open-vocabulary prompt-tuning pipelines, GPMN explicitly targets both base-class bias and seen–unseen overlap at the region level. Extensive experiments on VOC10-5-5 and VOC-COCO benchmarks demonstrate that GPMN consistently improves unknown recall and few-shot mAP over representative FS-OSOD baselines. These results suggest that prompt-level decoupling mitigates base-class bias, whereas memory-anchored regularization enlarges the seen–unseen margin, jointly supporting reliable unknown rejection in scarce-supervision regimes. Full article

Distributed optical fiber vibration sensing (DVS) has shown strong potential in perimeter security, pipeline leakage monitoring, transportation safety, and structural health diagnostics owing to its high sensitivity, long-range coverage, and immunity to electromagnetic interference. However, severe cross-scene distribution mismatch is often encountered in real-world deployments: indoor, outdoor, and pipeline environments exhibit markedly different noise patterns and time–frequency characteristics, thereby degrading the generalization ability of models trained in a single scene. To address this challenge, we propose a Progressive Multi-Strategy Domain Alignment (PMSDA) framework for label-disjoint cross-scene vibration recognition. PMSDA uses a compact expansion–compression encoder together with complementary alignment mechanisms—maximum mean discrepancy (MMD), correlation alignment (CORAL), and adversarial domain discrimination—to learn a scene-robust latent space from a labeled indoor source and two unlabeled target domains (outdoor and pipeline) within a single alternating-training model. Because the fine-grained source and target label spaces are disjoint, PMSDA is formulated as a representation-transfer framework rather than a standard label-shared unsupervised domain adaptation method; target-domain recognition is therefore performed through domain-specific prototype clustering in the aligned latent space. On three representative scenes with nine event classes in total, PMSDA achieved 89.5% accuracy, 86.7% macro-F1, and 0.93 AUC for Indoor→Outdoor, and 85.8%, 84.7%, and 0.87, respectively, for Indoor→Pipeline, outperforming traditional feature+SVM/RF pipelines, CNN/ResNet baselines, and representation-transfer baselines adapted from DANN/CDAN/SHOT under the same evaluation protocol. These results indicate that PMSDA is a promising and effective framework for offline cross-scene DVS evaluation under disjoint target event sets. Full article

Existing research has demonstrated that meta-learning methods hold considerable promise in addressing the challenges posed by few-shot object detection. However, remote sensing scenarios present two major challenges. The sparse features of small objects provide insufficient support information for query enhancement, and significant morphological variations caused by lighting and viewpoint differences hinder intra-class consistency capture via direct alignment in few-shot learning. To address these challenges, we propose a generative meta-learning detection framework. The framework first introduces a Dynamic Relation Dual-Stream Network to achieve dynamic support-query feature alignment through joint modeling of evolutionary and relational features, thereby enhancing representation in few-shot conditions. Second, an Optimal Transport-based Generative Meta-Learner is developed to mitigate feature distribution bias via generative augmentation in latent space. Additionally, an Orthogonal Frequency Decomposition Head is incorporated to adaptively separate query features into low-frequency contour and high-frequency detail components, effectively suppressing background noise interference. Experiments on multiple remote sensing datasets demonstrate that the proposed method achieves consistent performance gains over leading baseline methods in various few-shot settings. Its effectiveness is further validated across different backbone networks, highlighting strong generalization in few-shot remote sensing object detection. Full article

Stance detection has emerged as an essential tool in natural language processing for understanding how individuals express agreement, disagreement, or neutrality toward specific targets in social and online discourse. It plays a crucial role in bilingual and multilingual environments, including English-Arabic social media ecosystems, where differences in language structure, discourse style, and data availability pose significant challenges for reliable stance modelling. Existing approaches often struggle with target awareness, cross-lingual generalization, robustness to noisy user-generated text, and the interpretability of model decisions. This study aims to build a reliable, explainable target-aware bilingual stance-detection framework that generalizes across heterogeneous stance formats and languages without retraining on a dataset specific to the target language. Thus, a unified dual-encoder architecture based on mDeBERTa-v3 is proposed. Cross-language contrastive learning offers an auxiliary training objective to align English and Arabic stance representations in a common semantic space. Robustness-oriented regularization is used to mitigate the effects of informal language, vocabulary variation, and adversarial noise. To promote transparency and trustworthiness, the framework incorporates token-level rationale extraction, enables fine-grained interpretability, and supports analysis of hallucination. The proposed model is tested on a combined bilingual test set and two structurally distinct zero-shot benchmarks: MT-CSD and AraStance. Experimental results show consistent performance, with accuracies of 85.0% and 86.8% and F1-scores of 84.7% and 86.8% on the zero-shot benchmarks, confirming stable performance and realistic generalization. Ultimately, these findings reveal that effective bilingual stance detection can be achieved via explicit target conditioning, cross-lingual alignment, and explainability-driven design. Full article

Few-shot object detection in remote sensing imagery faces significant challenges, including limited labeled samples, complex scene backgrounds, and subtle inter-class differences. To tackle these issues, we design a novel detection framework that effectively transfers supervision from a few annotated support examples to the query domain. We introduce a feature enhancement mechanism that injects fine-grained support cues into the query representation, helping the model focus on relevant regions and suppress background noise. This allows the model to generate more accurate proposals and perform robust classification, especially for visually confusing or small objects. Additionally, our method enhances feature interaction between support and query images through a nonlinear combination strategy, which captures both semantic similarity and discriminative differences. The proposed framework is fully end-to-end and jointly optimizes the feature fusion and detection processes. Experiments on three challenging benchmarks, NWPU VHR-10, iSAID and DIOR, demonstrate that our method consistently achieves state-of-the-art results under different few-shot settings and category splits. Compared with other advanced methods, it yields superior performance, highlighting its strong generalization ability in low-data remote sensing scenarios. Full article

Object detection in adverse weather remains challenging due to the simultaneous degradation of visibility, structural boundaries, and semantic consistency. Existing restoration-driven or multi-branch detection approaches often fail to recover task-relevant features or introduce substantial computational overhead. To address this problem, DLC-SSD, a lightweight degradation-aware framework for detecting robust objects in adverse weather environments, is proposed. The framework integrates image enhancement and feature refinement into a single detection pipeline and adopts a hierarchical strategy in which global and local degradations are corrected at the image level, structural cues are reinforced in shallow high-resolution features, and semantic representations are refined in deep layers to suppress weather-induced noise. These components are jointly optimized end-to-end with the single-shot multibox detection (SSD) backbone. In rain, fog, and low-light conditions, DLC-SSD demonstrated more stable performance than conventional detectors and maintained a quasi-real-time inference speed, confirming its practicality in intelligent monitoring and autonomous driving environments. Full article

High-voltage direct-current (HVDC) systems are essential for large-scale renewable integration and asynchronous interconnection, yet their complex topologies and multi-type faults expose the limits of threshold- and signal-based diagnostics. These methods degrade under noisy, heterogeneous measurements acquired under dynamic operating conditions, resulting in poor adaptability, reduced accuracy, and high latency. To overcome these shortcomings, the synergistic use of knowledge graphs (KGs) and pre-trained models (PTMs) is emerging as a next-generation paradigm. KGs encode equipment parameters, protection logic, and fault propagation paths in an explicit, human-readable structure, while PTMs provide transferable representations that remain effective under label scarcity and data diversity. Coupled within a perception–cognition–decision loop, PTMs first extract latent fault signatures from multi-modal records; KGs then enable interpretable causal inference, yielding both precise localization and transparent explanations. This work systematically reviews the theoretical foundations, fusion strategies, and implementation pipelines of KG-PTM frameworks tailored to HVDC systems, benchmarking them against traditional diagnostic schemes. The paradigm demonstrates superior noise robustness, few-shot generalization, and decision explainability. However, open challenges remain, such as automated, conflict-free knowledge updating; principled integration of electro-magnetic physical constraints; real-time, resource-constrained deployment; and quantifiable trustworthiness. Future research should therefore advance autonomous knowledge engineering, physics-informed pre-training, lightweight model compression, and standardized evaluation platforms to translate KG-PTM prototypes into dependable industrial tools for intelligent HVDC operation and maintenance. Full article

Industrial visual inspection demands high-precision anomaly detection amid scarce annotations and unseen defects. This paper introduces a zero-shot framework leveraging multimodal feature fusion and stabilized attention pooling. CLIP’s global semantic embeddings are hierarchically aligned with DINOv2’s multi-scale structural features via a Dual-Modality Attention (DMA) mechanism, enabling effective cross-modal knowledge transfer for capturing macro- and micro-anomalies. A Stabilized Attention-based Pooling (SAP) module adaptively aggregates discriminative representations using self-generated anomaly heatmaps, enhancing localization accuracy and mitigating feature dilution. Trained solely in auxiliary datasets with multi-task segmentation and contrastive losses, the approach requires no target-domain samples. Extensive evaluation across seven benchmarks (MVTec AD, VisA, BTAD, MPDD, KSDD, DAGM, DTD-Synthetic) demonstrates state-of-the-art performance, achieving 93.4% image-level AUROC, 94.3% AP, 96.9% pixel-level AUROC, and 92.4% AUPRO on average. Ablation studies confirm the efficacy of DMA and SAP, while qualitative results highlight superior boundary precision and noise suppression. The framework offers a scalable, annotation-efficient solution for real-world industrial anomaly detection. Full article

Deep learning-based synthetic aperture radar (SAR) target recognition often suffers from overfitting under few-shot conditions, making it difficult to fully exploit the discriminative features contained in limited samples. Moreover, SAR targets frequently exhibit highly similar background scattering patterns, which further increase intra-class variations and reduce inter-class separability, thereby constraining the performance of few-shot recognition. To address these challenges, this paper proposes an adaptive contrastive metric (ACM) network with background suppression for few-shot SAR target recognition. Specifically, a spatial squeeze-and-excitation (SSE) attention module is introduced to adaptively highlight salient scattering structures of the target while effectively suppressing noise and irrelevant background interference, thus enhancing the robustness of feature representation. In addition, an ACM module is designed, where query samples are compared not only with their corresponding support class but also with the remaining classes. This enables explicit suppression of confusing background features and enlarges inter-class margins, thereby improving the discriminability of the learned feature space. The experimental results on publicly available SAR target recognition datasets demonstrate that the proposed method achieves significant improvements in background suppression and consistently outperforms several state-of-the-art metric-based few-shot learning approaches, validating the effectiveness and generalizability of the proposed framework. Full article

This study introduces a physics-guided self-supervised framework for few-shot ultrasonic defect detection in concrete structures, addressing the dual challenges of scarce labels and domain variability in structural health monitoring (SHM). Our method integrates physics-informed augmentations, contrastive representation learning, and adversarial domain alignment within a mutually reinforcing cycle, enabling robust defect classification with minimal supervision. A Physics-Informed Augmentation Module synthesizes realistic ultrasonic signals, training a Transformer encoder to extract invariant features while suppressing sensor noise. An Adversarial Feature Aligner further improves cross-domain generalization by mitigating distribution shifts across heterogeneous concretes. Experimental validation on three benchmark datasets demonstrates 63–66% accuracy in one-shot cross-domain tasks and up to 89% in five-shot settings. These results represent 12–15 percentage point gains over modern few-shot baselines, with improvements statistically significant at p < 0.001. Compatible with existing ultrasonic hardware, the proposed framework bridges physics-based modeling and machine learning while paving the way for scalable, field-ready SHM solutions for aging infrastructure and resilient smart cities. Full article

Understanding how the human brain processes and interprets multimedia content represents a frontier challenge in neuroscience and artificial intelligence. This study introduces a novel approach to decode semantic information from electroencephalogram (EEG) signals recorded during visual stimulus perception. We present DCT-ViT, a spatial–temporal transformer architecture that pioneers automated semantic recognition from brain activity patterns, advancing beyond conventional brain state classification to interpret higher level cognitive understanding. Our methodology addresses three fundamental innovations: First, we develop a topology-preserving 2D electrode mapping that, combined with temporal indexing, generates 3D spatial–temporal representations capturing both anatomical relationships and dynamic neural correlations. Second, we integrate discrete cosine transform (DCT) embeddings with standard patch and positional embeddings in the transformer architecture, enabling frequency-domain analysis that quantifies activation variability across spectral bands and enhances attention mechanisms. Third, we introduce the Semantics-EEG dataset comprising ten semantic categories extracted from visual stimuli, providing a benchmark for brain-perceived semantic recognition research. The proposed DCT-ViT model achieves 72.28% recognition accuracy on Semantics-EEG, substantially outperforming LSTM-based and attention-augmented recurrent baselines. Ablation studies demonstrate that DCT embeddings contribute meaningfully to model performance, validating their effectiveness in capturing frequency-specific neural signatures. Interpretability analyses reveal neurobiologically plausible attention patterns, with visual semantics activating occipital–parietal regions and abstract concepts engaging frontal–temporal networks, consistent with established cognitive neuroscience models. To address systematic misclassification between perceptually similar categories, we develop a hierarchical classification framework with boundary refinement mechanisms. This approach substantially reduces confusion between overlapping semantic categories, elevating overall accuracy to 76.15%. Robustness evaluations demonstrate superior noise resilience, effective cross-subject generalization, and few-shot transfer capabilities to novel categories. This work establishes the technical foundation for brain–computer interfaces capable of decoding semantic understanding, with implications for assistive technologies, cognitive assessment, and human–AI interaction. Both the Semantics-EEG dataset and DCT-ViT implementation are publicly released to facilitate reproducibility and advance research in neural semantic decoding. Full article

Few-Shot Industrial Anomaly Detection (FSIAD) is an essential yet challenging problem in practical scenarios such as industrial quality inspection. Its objective is to identify previously unseen anomalous regions using only a limited number of normal support images from the same category. Recently, large pre-trained vision-language models (VLMs), such as CLIP, have exhibited remarkable few-shot image-text representation abilities across a range of visual tasks, including anomaly detection. Despite their promise, real-world industrial anomaly datasets often contain noisy labels, which can degrade prompt learning and detection performance. In this paper, we propose AnomalyNLP, a new Noisy-Label Prompt Learning approach designed to tackle the challenge of few-shot anomaly detection. This framework offers a simple and efficient approach that leverages the expressive representations and precise alignment capabilities of VLMs for industrial anomaly detection. First, we design a Noisy-Label Prompt Learning (NLPL) strategy. This strategy utilizes feature learning principles to suppress the influence of noisy samples via Mean Absolute Error (MAE) loss, thereby improving the signal-to-noise ratio and enhancing overall model robustness. Furthermore, we introduce a prompt-driven optimal transport feature purification method to accurately partition datasets into clean and noisy subsets. For both image-level and pixel-level anomaly detection, AnomalyNLP achieves state-of-the-art performance across various few-shot settings on the MVTecAD and VisA public datasets. Qualitative and quantitative results on two datasets demonstrate that our method achieves the largest average AUC improvement over baseline methods across 1-, 2-, and 4-shot settings, with gains of up to 10.60%, 10.11%, and 9.55% in practical anomaly detection scenarios. Full article

Few-shot image classification seeks to recognize novel categories from only a handful of labeled examples, but conventional metric-based methods that rely mainly on global image features often produce unstable prototypes under extreme data scarcity, while local-descriptor approaches can lose context and suffer from inter-class local-pattern overlap. To address these limitations, we propose a Global–Local Feature Fusion network that combines a frozen, pretrained global feature branch with a self-attention based multi-local feature fusion branch. Multiple random crops are encoded by a shared backbone (ResNet-12), projected to Query/Key/Value embeddings, and fused via scaled dot-product self-attention to suppress background noise and highlight discriminative local cues. The fused local representation is concatenated with the global feature to form robust class prototypes used in a prototypical-network style classifier. On four benchmarks, our method achieves strong improvements: Mini-ImageNet 70.31% ± 0.20 (1-shot)/85.91% ± 0.13 (5-shot), Tiered-ImageNet 73.37% ± 0.22/87.62% ± 0.14, FC-100 47.01% ± 0.20/64.13% ± 0.19, and CUB-200-2011 82.80% ± 0.18/93.19% ± 0.09, demonstrating consistent gains over competitive baselines. Ablation studies show that (1) naive local averaging improves over global-only baselines, (2) self-attention fusion yields a large additional gain (e.g., +4.50% in 1-shot on Mini-ImageNet), and (3) concatenating global and fused local features gives the best overall performance. These results indicate that explicitly modeling inter-patch relations and fusing multi-granularity cues produces markedly more discriminative prototypes in few-shot regimes. Full article

The object detection model for synthetic aperture radar (SAR) images needs to have strong generalization ability and more stable detection performance due to the complex scattering mechanism, high sensitivity of the orientation angle, and susceptibility to speckle noise. Visual large models possess strong generalization capabilities for natural image processing, but their application to SAR imagery remains relatively rare. This paper attempts to introduce a visual large model into the SAR object detection task, aiming to alleviate the problems of weak cross-domain generalization and poor adaptability to few-shot samples caused by the characteristics of SAR images in existing models. The proposed model comprises an image encoder, an attention module, and a detection decoder. The image encoder leverages the pre-trained Segment Anything Model (SAM) for effective feature extraction from SAR images. An Adaptive Channel Interactive Attention (ACIA) module is introduced to suppress SAR speckle noise. Further, a Dynamic Tandem Attention (DTA) mechanism is proposed in the decoder to integrate scale perception, spatial focusing, and task adaptation, while decoupling classification from detection for improved accuracy. Leveraging the strong representational and few-shot adaptation capabilities of large pre-trained models, this study evaluates their cross-domain and few-shot detection performance on SAR imagery. For cross-domain detection, the model was trained on AIR-SARShip-1.0 and tested on SSDD, achieving an mAP50 of 0.54. For few-shot detection on SAR-AIRcraft-1.0, using only 10% of the training samples, the model reached an mAP50 of 0.503. Full article