Search Results (338)

Person re-identification (re-ID) aims to retrieve images of a given individual from different camera views. Obstacles obstructing parts of a pedestrian’s body often result in incomplete identity information, impairing recognition performance. To address the occlusion problem, a method called Multi-Branch Interaction Network (MBIN) is proposed, which exploits the information interaction between different branches to effectively characterize occluded pedestrians for person re-ID. The method consists primarily of a hard branch, a soft branch, and a view branch. The hard branch enhances feature robustness via a unified horizontal partitioning strategy. The soft branch improves the high-level feature representation via multi-head attention. The view branch fuses multi-view feature maps to form a comprehensive representation via a dual-classifier fusion mechanism. Moreover, a mutual knowledge distillation strategy is employed to promote knowledge exchange among the three branches. Extensive experiments are conducted on widely used person re-ID datasets to validate the effectiveness of our method. Full article

The objective of person re-identification (Re-ID) is to recognize a given target pedestrian across different cameras. However, perspective variations, resulting from differences in shooting angles, often significantly impact the accuracy of person Re-ID. To address this issue, this paper presents an attention-enhanced person Re-ID algorithm based on appearance–gait information interaction. Specifically, appearance features and gait features are first extracted from RGB images and gait energy images (GEIs), respectively, using two ResNet-50 networks. Then, a multimodal information exchange module based on the attention mechanism is designed to build a bridge for information exchange between the two modalities during the feature extraction process. This module aims to enhance the feature extraction ability through mutual guidance and reinforcement between the two modalities, thereby improving the model’s effectiveness in integrating the two types of modal information. Subsequently, to further balance the signal-to-noise ratio, importance weight estimation is employed to map perspective information into the importance weights of the two features. Finally, based on the autoencoder structure, the two features are weighted and fused under the guidance of importance weights to generate fused features that are robust to perspective changes. The experimental results on the CASIA-B dataset indicate that, under conditions of viewpoint variation, the method proposed in this paper achieved an average accuracy of 94.9%, which is 1.1% higher than the next best method, and obtained the smallest variance of 4.199, suggesting that the method proposed in this paper is not only more accurate but also more stable. Full article

Background: Gingival recession (GR) is defined as the exposure of the root surface due to the gingival margin shifting apically from the cemento-enamel junction. Current effective management of defects related to GR relies on root coverage periodontal plastic surgery (RCPPS), using the Modified Coronally Advanced Flap (mCAF) with an envelope design. Recent literature also reported the association of different biomaterials to the mCAF procedure. In light of these considerations, a systematic review (SR) was conducted to determine and compare the efficacy of all mCAF adjunctive techniques for the treatment of multiple adjacent GR-type (MAGR) defects. Methods: An electronic search was conducted in 2025 on studies published between 2013 and 2025, using PubMed, Scopus, Web of Science, and Cinahl Complete, to address the focused question: “which is the efficacy of different mCAF adjunctive techniques for the treatment of multiple adjacent GR-type defects, in terms of root coverage (RC), esthetic outcomes, and keratinized tissue (KT) augmentation?”. Randomized controlled trials with a minimum follow-up of 6 months with ≥ 5 patients treated for coverage of MAGR were included. Risk of bias was assessed with RoB 2 Tool. A meta-analysis was performed using RevMan5.4 software and the level of evidence of included studies was analyzed with GRADEPro GDT. Results: A total of 17 studies were included in the SR, 9 of which evaluating mCAF + sCTG (subepithelial connective tissue graft) vs. mCAF adjunctive techniques [Collagen Matrix (CM), xenogeneic acellular dermal matrix (XADM), Platelet-Rich Fibrin (PRF), Enamel Matrix Derivatives (EMD), sCTG harvested double blade scalpel] were then included in the meta-analysis. The primary outcomes of complete root coverage (CRC) and keratinized tissue width variation (ΔKTW) were statistically significant ([CRC: Odds Ratio (OR) 1.70; 95% CI (confidence interval) 1.18, 2.44; p = 0.004]; [ΔKTW: SMD (standardized mean difference) 0.37; 95% CI 0.11, 0.63; p = 0.005]) in favor of mCAF + CTG. Meanwhile, no statistically significant difference was observed in terms of RES. The certainty assessment highlighted relevant results: despite the lack of evidence in the long-term, a high level of evidence showed that sCTG was more effective than XADM in terms of CRC (p = 0.002) and ΔKTW (p = 0.0001). A low level of evidence revealed that sCTG achieved a greater ΔKTW compared to CM (p = 0.0006). Although no significant differences were observed, a low level of evidence suggested that mCAF + EMD and mCAF + sCTG (DBS) may provide good results. To date, only one RCT showed long-term stable results of CTG in terms of RC. Conclusions: The association of sCTG to mCAF demonstrated better results in terms of RC and KTW augmentation in short- and medium-term follow-ups. Long-term studies are needed to confirm the efficacy of the other mCAF adjunctive techniques, considering limitations due to heterogeneity in follow-ups, distribution of techniques analyzed, and different study designs. Registration in PROSPERO (International prospective register of systematic reviews) was performed with ID CRD420251085823. Full article

Background: Kidney transplant recipients (KTRs) experience improved survival and quality of life compared to dialysis treatment, but chronic immunosuppression may increase the risk of de novo post-transplant cancer. Colorectal cancer (CRC) is increasingly recognized in this population. This systematic review aims to synthetize contemporary evidence on CRC epidemiology, outcomes, and risk determinants among KTRs. Methods: A comprehensive search for observational and registry-based studies reporting CRC in adult KTRs was conducted on PubMed, Scopus, Web of Science, and ProQuest. The studies found were subsequently subjected to screening, data extraction, and the risk-of-bias appraisal process. Due to heterogeneity, findings were synthesized narratively. Results: Twenty-six studies encompassing 863,005 KTRs met inclusion criteria: 22 retrospective cohorts, 1 prospective cohort, 2 cross-sectional, and 1 case-control. Absolute CRC occurrence varies by geography, population, and follow-up. Reported risks ranged from no excess to modestly elevated standardized incidence ratios (SIRs): ~0.76–3.60 overall, with a right-sided colon predominance. Overall, higher mortality and worse prognosis were reported in kidney transplant recipients with colorectal cancer compared to the general population, as a result of later-stage diagnosis and more aggressive histologies. Consistent risk factors included older age, time since transplantation, diabetes and metabolic comorbidities, prior dialysis duration/graft failure, and smoking; the female sex showed higher relative CRC risk in some cohorts. The remarkable role of immunosuppression profiles was consistently highlighted: cyclosporine—azathioprine maintenance and alemtuzumab induction were associated with higher CRC risk in large registries, whereas tacrolimus—mycophenolate regimens showed lower risk signals and mTOR inhibitors suggested possible protective effects. Conclusions: Contemporary evidence suggests a modest, heterogenous increase in CRC risk among KTRs, a proximal (right-sided) predominance, and a tendency toward advanced-stage presentation with reduced survival. These findings justify the need to consider risk-tailored, lifelong surveillance strategies anchored in a full colonoscopy, with earlier initiation in younger or otherwise high-risk recipients, alongside careful optimization and periodic re-evaluation of immunosuppression. Prospective multicenter studies and cost-effectiveness analyses should refine screening thresholds and therapeutic strategies. PROSPERO ID: CRD420251071658. Full article

Background/Objectives: X-linked intellectual disability (XLID) is a highly heterogeneous disorder accounting for ~10% of all males with ID. Next-generation sequencing (NGS) has revolutionized the discovery of causal XLID genes and variants; however, many cases remain unresolved. We present a four-generation syndromic XLID family with multiple males exhibiting variable degree of ID, focal dystonia and epilepsy. Methods: Extensive cytogenetic and targeted genetic testing was initially performed, followed by whole-exome sequencing (WES) and short-read whole-genome sequencing (WGS). Apart from the routine NGS analysis pipelines, sequencing data was revisited by focusing on poorly covered/mapped regions on chromosome X (chrX), to potentially reveal unidentified clinically relevant variants. Candidate variant validation and family segregation analysis were performed with Sanger sequencing. Results: All initial diagnostic testing was negative. Subsequently, 300 previously reported “dark” chrX coding DNA sequences, overlapping 97 genes, were cross-checked against 29 chrX genes highly associated (p < 0.05) with ID and focal dystonia, according to Phenomizer. Manual inspection of the existing NGS data in two low-coverage regions, chrX:25013469-25013696 and chrX:111744737-111744820 (hg38), revealed a recurrent pathogenic ARX variant NM_139058.3:c.441_464dup p.(Ala148_Ala155dup) (ARXdup24) associated with non-syndromic or syndromic XLID, including Partington syndrome. Sanger sequencing confirmed ARXdup24 in all affected males, with carrier status in their unaffected mothers, and absence in other unaffected relatives. Conclusions: After several years of diagnostic odyssey, the pathogenic ARXdup24 variant was unmasked, supporting a genotype–phenotype correlation in the first Partington syndrome family in Cyprus. This study highlights that re-examining underrepresented genomic regions and using phenotype-driven tools can provide critical diagnostic insights in unresolved XLID cases. Full article

Northern corn leaf blight (NCLB), caused by the fungus Exserohilum turcicum, is one of the most significant foliar diseases in maize worldwide, with its severity being highly influenced by environmental conditions. An effective strategy used to control NCLB involves screening diverse maize germplasm for resistant sources through multi-environment inoculation assays, ultimately aiming to develop resistant varieties. This study systematically evaluated 711 maize germplasm accessions with rich genetic diversity. The evaluation was conducted under four location–year environment combinations (Shangluo, Shaanxi Province, China in 2014–2015 and Xinzhou, Shanxi Province, China in 2021–2022) using artificial inoculation with physiological race 123N (or races 1, 2, 3, N). The results showed that the estimated variances of genotype, environment, and genotype-by-environment interaction were all highly significant (p < 0.01). Significant correlations (p < 0.01) were observed among replicates within each environment, with correlation coefficients (r) ranging from 0.67 to 0.88. At the Xinzhou trial in 2021, four replicates were inoculated with four physiological races (1, 2, 3, and N), revealing highly significant correlations (r = 0.77–0.80, p < 0.01) among them. The disease severity of the tropical germplasm was significantly lower (p < 0.001) than that of the temperate germplasm. Among the temperate subgroups, the PA and PB (groups A and B germplasms derived from modern US hybrids) subgroups exhibited lower disease severity, with the PB subgroup showing the lowest, while the Iodent and Reid subgroups exhibited higher susceptibility. The disease severity responses to the four physiological races were highly positively correlated (r = 0.77–0.80, p < 0.001), and their correlations with the composite inoculation (race 123N) ranged from 0.65 to 0.83. Based on the resistance evaluations across four location–year environment combinations, the 711 maize accessions were classified into five categories: 20 were highly resistant, 236 resistant, 205 moderately resistant, 237 susceptible, and 13 highly susceptible. The findings indicate that the tropical germplasm and the temperate PB subgroup are major sources of NCLB resistance. Full article

Occluded person re-identification (Re-ID) faces significant challenges, mainly due to the interference of occlusion noise and the scarcity of realistic occluded training data. Although data augmentation is a commonly used solution, the current occlusion augmentation methods suffer from the problem of dual inconsistencies: intra-sample inconsistency is caused by misaligned synthetic occluders (an augmentation operation for simulating real occlusion situations); i.e., randomly pasted occluders ignore spatial prior information and style differences, resulting in unrealistic artifacts that mislead feature learning; inter-sample inconsistency stems from information loss during random cropping (an augmentation operation for simulating occlusion-induced information loss); i.e., single-scale cropping strategies discard discriminative regions, weakening the robustness of the model. To address the aforementioned dual inconsistencies, this study proposes the unified Multi-Aligned and Multi-Scale Augmentation (MA–MSA) framework based on the core principle of ”synthetic data should resemble real-world data”. First, the Frequency–Style–Position Data Augmentation (FSPDA) module is designed: it ensures consistency in three aspects (frequency, style, and position) by constructing an occluder library that conforms to real-world distribution, achieving style alignment via adaptive instance normalization and optimizing the placement of occluders using hierarchical position rules. Second, the Multi-Scale Crop Data Augmentation (MSCDA) strategy is proposed. It eliminates the problem of information loss through multi-scale cropping with non-overlapping ratios and dynamic view fusion. In addition, different from the traditional serial augmentation method, MA–MSA integrates FSPDA and MSCDA in a parallel manner to achieve the collaborative resolution of dual inconsistencies. Extensive experiments on Occluded-Duke and Occluded-REID show that MA–MSA achieves state-leading performance of 73.3% Rank-1 (+1.5%) and 62.9% mAP on Occluded-Duke, and 87.3% Rank-1 (+2.0%) and 82.1% mAP on Occluded-REID, demonstrating superior robustness without auxiliary models. Full article

Background: Personalizing adjuvant chemotherapy (ACT) after curative resection in early-stage NSCLC remains unmet because prior ACT-biomarker findings rarely reproduce across studies. Key barriers are platform and preprocessing heterogeneity, dominant batch effects, and incomplete ACT annotations. As a result, many signatures that perform well in a single cohort fail during external validation. We created an open, harmonized meta-database linking gene expression with curated ACT exposure and survival to enable fair benchmarking and modeling. Methods: A PRISMA-guided search of 999 GEO studies (through January 2025) used LLM-assisted triage of titles, clinical tables, and free text to identify datasets with explicit ACT status and patient-level survival. Eight Affymetrix microarray cohorts (GPL570/GPL96) met eligibility. Raw CEL files underwent robust multi-array average; probes were re-annotated to Entrez IDs and collapsed by median. Covariate-preserving ComBat adjusted platform/study while retaining several clinical factors. Batch structure was quantified by principal-component analysis (PCA) variance, silhouette width, and UMAP. Two quality-control (QC) filters, median M-score deviation and PCA leverage, flagged and removed technical outliers. Results: The final meta-database comprises 1340 patients (223 (16.6%) ACT; 1117 (83.4%) observation), 13,039 intersecting genes, and 594 overall-survival events. Batch-associated variance (PC1 + PC2) decreased from 63.1% to 20.1%, and mean silhouette width shifted from 0.82 to −0.19 post-correction. Seven arrays (0.5%) were excluded by QC. Event depth supports high-dimensional survival and heterogeneity-of-treatment modeling, and the multi-cohort design enables internal–external validation. Conclusions: This first open, rigorously harmonized NSCLC transcriptomic database provides the sample size, demographic diversity, and technical consistency required to benchmark ACT-benefit markers. By making these data openly available, it will accelerate equitable precision-oncology research and enable data-driven treatment decisions in early-stage NSCLC. Full article

Sketch re-identification (Re-ID) aims to retrieve pedestrian photographs in the gallery dataset by a query sketch image drawn by professionals, which is crucial for criminal investigations and missing person searches in the field of public security. The main challenge of this task lies in bridging the significant modality gap between sketches and photos while extracting discriminative modality-invariant features. However, information asymmetry between sketches and RGB photographs, particularly the differences in color information, severely interferes with cross-modal matching processes. To address this challenge, we propose a novel network architecture that integrates multi-channel augmentation with hierarchically consistent disentanglement learning. Specifically, a multi-channel augmentation module is developed to mitigate the interference of color bias in cross-modal matching. Furthermore, a modality-disentangled prototype(MDP) module is introduced to decompose pedestrian representations at the feature level into modality-invariant structural prototypes and modality-specific appearance prototypes. Additionally, a cross-layer decoupling consistency constraint is designed to ensure the semantic coherence of disentangled prototypes across different network layers and to improve the stability of the whole decoupling process. Extensive experimental results on two public datasets demonstrate the superiority of our proposed approach over state-of-the-art methods. Full article

The increasing adoption of electric vehicles (EVs) within intelligent transportation systems (ITSs) has elevated the importance of cybersecurity, especially with the rise in Vehicle-to-Everything (V2X) communications. Traditional intrusion detection systems (IDSs) struggle to address the evolving and complex nature of cyberattacks in such dynamic environments. To address these challenges, this study introduces a novel deep learning-based IDS designed specifically for EV communication networks. We present a hybrid model that integrates convolutional neural networks (CNNs), long short-term memory (LSTM) layers, and adaptive learning strategies. The model was trained and validated using the VeReMi dataset, which simulates a wide range of attack scenarios in V2X networks. Additionally, an ablation study was conducted to isolate the contribution of each of its modules. The model demonstrated strong performance with 98.73% accuracy, 97.88% precision, 98.91% sensitivity, and 98.55% specificity, as well as an F1-score of 98.39%, an MCC of 0.964, a false-positive rate of 1.45%, and a false-negative rate of 1.09%, with a detection latency of 28 ms and an AUC-ROC of 0.994. Specifically, this work fills a clear gap in the existing V2X intrusion detection literature—namely, the lack of scalable, adaptive, and low-latency IDS solutions for hardware-constrained EV platforms—by proposing a hybrid CNN–LSTM architecture coupled with an elastic weight consolidation (EWC)-based adaptive learning module that enables online updates without full retraining. The proposed model provides a real-time, adaptive, and high-precision IDS for EV networks, supporting safer and more resilient ITS infrastructures. Full article

Local Feature generation for vehicle re-identification is a challenging research area that is not yet well-investigated. The part-based convolutional baseline model with refined part pooling (PCB-RPP) architecture commonly approached in person reidentification problems was experimented over two standard vehicle image datasets (VReId and VehicleId) to establish that RPP over uniform partitions do not work well. To address the limitation, we propose a novel approach, Overlapped-PCB, which overlaps parts of two adjacent parts to generate new parts to train the classifiers. The results are concatenated to generate the feature set and this improves the re-identification accuracy in comparison to the RPP approach. Performance comparison results of extensive testing are also presented using re-ranking and ensembling in the evaluation stage. Our proposed model has been ensembled over three architectures, ResNet50, ResNet101, and ResNext50, to show the extent of performance improvement over existing works. The re-ranking process is shown to be strongly dataset-dependent for which the conventionally used k-reciprocal neighbors method has been improved by augmenting a new simple score-based algorithm for obtaining the best mix of component distances. This can be used as a generalized tool to finetune re-ranking for different datasets. Full article

Person re-identification aims to match images of the same individual across non-overlapping cameras by analyzing personal characteristics. Recently, Transformer-based models have demonstrated excellent capabilities and achieved breakthrough progress in this task. However, their high computational costs and inadequate capacity to capture fine-grained local features impose significant constraints on re-identification performance. To address these challenges, this paper proposes a novel Toward Efficient Transformer-based Person Re-identification (TE-TransReID) framework. Specifically, the proposed framework retains only the former L-th layer layers of a pretrained Vision Transformer (ViT) for global feature extraction while combining local features extracted from a pretrained CNN, thus achieving the trade-off between high accuracy and lightweight networks. Additionally, we propose a dual efficient feature-fusion strategy to integrate global and local features for accurate person re-identification. The Efficient Token-based Feature-Fusion Module (ETFFM) employs the gate-based network to learn fused token-wise features, while the Efficient Patch-based Feature-Fusion Module (EPFFM) utilizes a lightweight Transformer to aggregate patch-level features. Finally, TE-TransReID achieves a rank-1 of 94.8%, 88.3%, and 85.7% on Market1501, DukeMTMC, and MSMT17 with a parameter of 27.5 M, respectively. Compared to existing CNN–Transformer hybrid models, TE-TransReID maintains comparable recognition accuracy while drastically reducing model parameters, establishing an optimal equilibrium between recognition accuracy and computational efficiency. Full article

Unmanned aerial vehicles (UAVs) have become indispensable tools for surveillance, enabled by their ability to capture multi-perspective imagery in dynamic environments. Among critical UAV-based tasks, cross-platform person search—detecting and identifying individuals across distributed camera networks—presents unique challenges. Severe viewpoint variations, occlusions, and cluttered backgrounds in UAV-captured data degrade the performance of conventional discriminative models, which struggle to maintain robustness under such geometric and semantic disparities. To address this, we propose view-invariant person search (VIPS), a novel two-stage framework combining Faster R-CNN with a view-invariant re-Identification (VIReID) module. Unlike conventional discriminative models, VIPS leverages the semantic flexibility of large vision–language models (VLMs) and adopts a two-stage training strategy to decouple and align text-based ID descriptors and visual features, enabling robust cross-view matching through shared semantic embeddings. To mitigate noise from occlusions and cluttered UAV-captured backgrounds, we introduce a learnable mask generator for feature purification. Furthermore, drawing from vision–language models, we design view prompts to explicitly encode perspective shifts into feature representations, enhancing adaptability to UAV-induced viewpoint changes. Extensive experiments on benchmark datasets demonstrate state-of-the-art performance, with ablation studies validating the efficacy of each component. Beyond technical advancements, this work highlights the potential of VLM-derived semantic alignment for UAV applications, offering insights for future research in real-time UAV-based surveillance systems. Full article

Driven by growing public security demands and the advancement of intelligent surveillance systems, person re-identification (ReID) has emerged as a prominent research focus in the field of computer vision. However, this task presents challenges due to its high sensitivity to variations in visual appearance caused by factors such as body pose and camera parameters. Although deep learning-based methods have achieved marked progress in ReID, the high cost of annotation remains a challenge that cannot be overlooked. To address this, we propose an unsupervised attribute learning framework that eliminates the need for costly manual annotations while maintaining high accuracy. The framework learns the mid-level human attributes (such as clothing type and gender) that are robust to substantial visual appearance variations and can hence boost the accuracy of attributes with a small amount of labeled data. To carry out our framework, we present a part-based convolutional neural network (CNN) architecture, which consists of two components for image and body attribute learning on a global level and upper- and lower-body image and attribute learning at a local level. The proposed architecture is trained to learn attribute-semantic and identity-discriminative feature representations simultaneously. For model learning, we first train our part-based network using a supervised approach on a labeled attribute dataset. Then, we apply an unsupervised clustering method to assign pseudo-labels to unlabeled images in a target dataset using our trained network. To improve feature compatibility, we introduce an attribute consistency scheme for unsupervised domain adaptation on this unlabeled target data. During training on the target dataset, we alternately perform three steps: extracting features with the updated model, assigning pseudo-labels to unlabeled images, and fine-tuning the model. Through a unified framework that fuses complementary attribute-label and identity label information, our approach achieves considerable improvements of 10.6% and 3.91% mAP on Market-1501→DukeMTMC-ReID and DukeMTMC-ReID→Market-1501 unsupervised domain adaptation tasks, respectively. Full article

Although SAR systems can provide high-resolution aerial view images all-day, all-weather, the aspect and pose-sensitivity of the SAR target signatures, which defies the Gestalt perceptual principles, sets a frustrating performance upper bound for SAR Automatic Target Recognition (ATR). Therefore, we propose a network to support context-guided ATR by using aligned Electro-Optical (EO)-SAR image pairs. To realize EO-SAR image scene grammar alignment, the stable context features highly correlated to the parking patterns of the vehicle and aircraft targets are extracted from the EO images as prior knowledge, which is used to assist SAR-ATR. The proposed network consists of a Scene Recognition Module (SRM) and an instance-level Cross-modality ATR Module (CATRM). The SRM is based on a novel light-condition-driven adaptive EO-SAR decision weighting scheme, and the Outlier Exposure (OE) approach is employed for SRM training to realize Out-of-Distribution (OOD) scene detection. Once the scene depicted in the cut of interest is identified with the SRM, the image cut is sent to the CATRM for ATR. Considering that the EO-SAR images acquired from diverse observation angles often feature unbalanced quality, a novel class-incremental learning method based on the Context-Guided Re-Identification (ReID)-based Key-view (CGRID-Key) exemplar selection strategy is devised so that the network is capable of continuous learning in the open-world deployment environment. Vehicle ATR experimental results based on the UNICORN dataset, which consists of 360-degree EO-SAR images of an army base, show that the CGRID-Key exemplar strategy offers a classification accuracy 29.3% higher than the baseline model for the incremental vehicle category, SUV. Moreover, aircraft ATR experimental results based on the aligned EO-SAR images collected over several representative airports and the Arizona aircraft boneyard show that the proposed network achieves an F1 score of 0.987, which is 9% higher than YOLOv8. Full article