Search Results (25)

Dental implant placement has evolved from conventional freehand techniques toward digitally guided workflows integrating cone-beam computed tomography (CBCT), computer-aided design/computer-aided manufacturing (CAD/CAM), and dynamic navigation systems. Although guided surgery improves positional accuracy, its clinical relevance compared with freehand placement remains debated. This narrative review, based on a systematic and structured literature search following predefined selection criteria, analyzes studies published between 2000 and 2025 comparing guided and freehand implant placement regarding accuracy, survival, complications, biological outcomes, and workflow efficiency. Searches of PubMed/MEDLINE, Embase, and Web of Science identified 40 eligible human clinical studies for qualitative synthesis. Guided placement consistently demonstrated greater positional accuracy, with angular deviations of approximately 2–4° versus 5–9° for freehand placement and linear deviations reduced by about 1 mm. Nevertheless, implant survival rates were high and comparable for both techniques, generally exceeding 95% across short- and medium-term follow-up. Overall complication rates were low; guided approaches reduced anatomical risk and improved prosthetic predictability in complex or multi-implant cases, while freehand placement allowed greater intraoperative flexibility and tactile feedback, potentially optimizing primary stability in variable bone conditions. Marginal bone loss and peri-implant tissue outcomes were similar between approaches. Guided workflows required additional planning time and costs but enhanced reproducibility in complex rehabilitations. Guided and freehand implant placement should therefore be considered complementary strategies, with optimal outcomes depending on case selection, surgical expertise, and the balanced integration of digital technologies into contemporary implant practice. Full article

Face recognition under uncontrolled lighting remains challenging due to variations in brightness, background noise, and low-quality features. This paper presents a unified deep learning model that integrates illumination normalization, skin-aware spatial modulation, and quality-based margin learning within a single inference process. Unlike earlier methods that treat relighting or segmentation as preprocessing, this approach directly integrates mask-guided feature modulation into embedding learning. The system comprises RetinaFace detection, photometric augmentation during training, lightweight neural relighting at inference, U-Net-based skin segmentation, and identity embeddings trained with ArcFace, AdaFace, or MagFace losses, with angular margins adapted to feature quality. Experiments on Labeled Faces in the Wild (LFW), Celebrities in Frontal-Profile (CFP-FP), Age Database 30 (AgeDB-30), and a custom illumination dataset demonstrate steady enhancements in difficult lighting conditions. The model reaches a competitive 99.8% accuracy on LFW and shows notable improvements on pose-hard CFP-FP and the custom dataset, such as a +2.6% increase in TPR at 1 ${\times 10}^{-4}$ FPR. The key innovations include: (i) mask-guided embedding modulation that embeds segmentation into feature learning, (ii) a dual strategy combining training-time photometric data augmentation with inference-time neural relighting, and (iii) joint spatial–quality margin learning via AdaFace/MagFace. Finally, results confirm consistent gains under challenging illumination and pose variations. Full article

Cross-age face analysis has emerged as an important topic in biometric recognition due to substantial facial appearance variations caused by aging. Nevertheless, most existing approaches primarily focus on face verification (1:1 matching) and frequently rely on explicit age annotations, which limit their applicability in large-scale retrieval scenarios. In this study, large-scale cross-age face retrieval (1:N matching) is investigated, and a Hybrid Metric Learning Framework (HMLF) is proposed to learn age-invariant and retrieval-oriented facial representations without requiring age labels. The proposed framework integrates Additive Angular Margin Loss (ArcFace) with supervised contrastive learning to enhance feature discriminability. Furthermore, a mixed triplet mining strategy is introduced to improve the effectiveness of hard sample selection. A memory bank-based InfoNCE formulation is incorporated to provide a large number of negative samples, and an uncertainty-based adaptive weighting scheme is designed to automatically balance multiple loss components during optimization. To better simulate realistic retrieval scenarios, an extended cross-age retrieval evaluation protocol is established. Extensive experimental results demonstrate that the proposed framework achieves superior retrieval performance across different backbone architectures. The results further provide systematic insights into the influence of backbone design, loss formulation, and optimization strategies on cross-age retrieval accuracy. Full article

Face recognition technology implemented on blockchain platforms enhances the security and integrity of face embeddings (the numerical representations extracted from facial images). However, it encounters unique privacy challenges due to the transparent and immutable nature of blockchains. Face embeddings hold sensitive biometric data that, once compromised, cannot be changed like conventional passwords. This study offers a new framework for using the Internet Computer Protocol (ICP), a decentralized blockchain platform, to implement CosPEEPChain (blockchain-secured privacy-preserving face recognition using eigenface perturbation and CosFace). CosPEEPChain integrates eigenface decomposition with local differential privacy (LDP) to ensure the privacy of face embeddings, CosFace for cosine margin learning’s discriminative ability on perturbed eigenface representations, and blockchain to ensure transparent and tamper-proof storage of face recognition models. We present CosPEEP (privacy-preserving face recognition using eigenface perturbation and CosFace), which shows substantial improvements and maintains consistent performance over baseline PEEP (privacy using eigenface perturbation), with a mean accuracy of 96.77 ± 0.85% and stability (std = 0.31–1.28%) across a range of privacy budgets ($ϵ\in [0.5,8.0]$) on the LFW dataset. Statistical significance testing confirms CosPEEP surpasses PEEP in 11/16 privacy budgets (p < 0.05) with a mean improvement of +1.92%. We also present ArcPEEP, which uses additive angular margin loss (ArcFace) to compare margin-based improvements. We verify the attributes of the models on the chain. In total, CosPEEPChain uses fewer cycles compared to the baseline ICP face recognition. Full article

Recent studies in Visual Question Answering (VQA) have revealed that models often rely heavily on language priors rather than vision–language understanding, leading to poor generalization under distribution shifts. To address this challenge, we propose HQD-EM, a unified debiasing framework that combines the Hierarchical Question Decomposition (HQD) module with an Ensemble adaptive angular Margin (EM) loss. HQD systematically decomposes questions into multi-granular representations to capture layered language biases, while EM leverages bias confidence to modulate per-sample decision margins dynamically. Our method integrates an ensemble-based method with adaptive margin learning in an end-to-end trainable architecture. Experiments on VQA benchmarks demonstrate that HQD-EM significantly outperforms prior works on VQA-CP2 and VQA-CP1. Full article

This paper proposes a multi-scale time-frequency representation fusion network (MTRFN) for target recognition in synthetic aperture radar (SAR) imagery. Leveraging the spectral characteristics of six radar sub-views, the model incorporates a multi-scale representation fusion (MRF) module to extract discriminative frequency-domain features from two types of radar sub-views with high learnability. Additionally, physical scattering characteristics in SAR images are captured via time-frequency domain analysis. To enhance feature integration, a gated fusion network performs adaptive feature concatenation. The MRF module integrates a lightweight residual block to reduce network complexity and employs a coordinate attention mechanism to prioritize salient targets in the frequency spectrum over background noise, aligning the model’s focus with physical scattering principles. Furthermore, the model introduces an angular additive margin loss function during classification to enhance intra-class compactness and inter-class separability while reducing computational overhead. Compared with existing interpretable methods, the proposed approach combines architectural transparency with physical interpretability, thereby lowering the risk of recognition errors. Extensive experiments conducted on four public datasets demonstrate that the proposed MTRFN significantly outperforms existing benchmark methods. Comparative experiments using heat maps further confirm that the proposed physical feature-guided module effectively directs the model’s attention toward the target rather than the background. Full article

Electoral fraud, particularly multiple voting, undermines the integrity of democratic processes. To address this challenge, this study introduces an innovative facial recognition system that integrates an enhanced 50-layer Residual Network (ResNet50) architecture with Additive Angular Margin Loss (ArcFace) and Multi-Task Cascaded Convolutional Neural Networks (MTCNN) for face detection. Using the Mahalanobis distance, the system verifies voter identities by comparing captured facial images with previously recorded biometric features. Extensive evaluations demonstrate the methodology’s effectiveness, achieving a facial recognition accuracy of 99.85%. This significant improvement over existing baseline methods has the potential to enhance electoral transparency and prevent multiple voting. The findings contribute to developing robust biometric-based electoral systems, thereby promoting democratic trust and accountability. Full article

The presence of non-tobacco-related materials can significantly compromise the quality of tobacco. To accurately detect non-tobacco-related materials, this study introduces a lightweight and real-time detection model derived from the YOLOv11 framework, named LRNTRM-YOLO. Initially, due to the sub-optimal accuracy in detecting diminutive non-tobacco-related materials, the model was augmented by incorporating an additional layer dedicated to enhancing the detection of small targets, thereby improving the overall accuracy. Furthermore, an attention mechanism was incorporated into the backbone network to focus on the features of the detection targets, thereby improving the detection efficacy of the model. Simultaneously, for the introduction of the SIoU loss function, the angular vector between the bounding box regressions was utilized to define the loss function, thus improving the training efficiency of the model. Following these enhancements, a channel pruning technique was employed to streamline the network, which not only reduced the parameter count but also expedited the inference process, yielding a more compact model for non-tobacco-related material detection. The experimental results on the NTRM dataset indicate that the LRNTRM-YOLO model achieved a mean average precision (mAP) of 92.9%, surpassing the baseline model by a margin of 4.8%. Additionally, there was a 68.3% reduction in the parameters and a 15.9% decrease in floating-point operations compared to the baseline model. Comparative analysis with prominent models confirmed the superiority of the proposed model in terms of its lightweight architecture, high accuracy, and real-time capabilities, thereby offering an innovative and practical solution for detecting non-tobacco-related materials in the future. Full article

Early detection of plant diseases is crucial for agro-holdings, farmers, and smallholders. Various neural network architectures and training methods have been employed to identify optimal solutions for plant disease classification. However, research applying one-shot or few-shot learning approaches, based on similarity determination, to the plantdisease classification domain remains limited. This study evaluates different loss functions used in similarity learning, including Contrastive, Triplet, Quadruplet, SphereFace, CosFace, and ArcFace, alongside various backbone networks, such as MobileNet, EfficientNet, ConvNeXt, and ResNeXt. Custom datasets of real-life images, comprising over 4000 samples across 68 classes of plant diseases, pests, and their effects, were utilized. The experiments evaluate standard transfer learning approaches alongside similarity learning methods based on two classes of loss function. Results demonstrate the superiority of cosine-based methods over Siamese networks in embedding extraction for disease classification. Effective approaches for model organization and training are determined. Additionally, the impact of data normalization is tested, and the generalization ability of the models is assessed using a special dataset consisting of 400 images of difficult-to-identify plant disease cases. Full article

Simultaneous localization and mapping, a critical technology for enabling the autonomous driving of vehicles and mobile robots, increasingly incorporates multi-sensor configurations. Inertial measurement units (IMUs), known for their ability to measure acceleration and angular velocity, are widely utilized for motion estimation due to their cost efficiency. However, the inherent noise in IMU measurements necessitates the integration of additional sensors to facilitate spatial understanding for mapping. Visual–inertial odometry (VIO) is a prominent approach that combines cameras with IMUs, offering high spatial resolution while maintaining cost-effectiveness. In this paper, we introduce our uncertainty-aware depth network (UD-Net), which is designed to estimate both depth and uncertainty maps. We propose a novel loss function for the training of UD-Net, and unreliable depth values are filtered out to improve VIO performance based on the uncertainty maps. Experiments were conducted on the KITTI dataset and our custom dataset acquired from various driving scenarios. Experimental results demonstrated that the proposed VIO algorithm based on UD-Net outperforms previous methods with a significant margin. Full article

Despite major breakthroughs in facial recognition technology, problems with bias and a lack of diversity still plague face recognition systems today. To address these issues, we created synthetic face data using a diffusion-based generative model and fine-tuned already-high-performing models. To achieve a more balanced overall performance across various races, the synthetic dataset was created by following the dual-condition face generator (DCFace) resolution and using race-varied data from BUPT-BalancedFace as well as FairFace. To verify the proposed method, we fine-tuned a pre-trained improved residual networks (IResnet)-100 model with additive angular margin (ArcFace) loss using the synthetic dataset. The results show that the racial gap in performance is reduced from 0.0107 to 0.0098 in standard deviation terms, while the overall accuracy increases from 96.125% to 96.1625%. The improved racial balance and diversity in the synthetic dataset led to an improvement in model fairness, demonstrating that this resource could facilitate more equitable face recognition systems. This method provides a low-cost way to address data diversity challenges and help make face recognition more accurate across different demographic groups. The results of the study highlighted that more advanced synthesized datasets, created through diffusion-based models, can also result in increased facial recognition accuracy with greater fairness, emphasizing that these should not be ignored by developers aiming to create artificial intelligence (AI) systems. Full article

With the continuous advancement of autonomous driving and monitoring technologies, there is increasing attention on non-intrusive target monitoring and recognition. This paper proposes an ArcFace SE-attention model-agnostic meta-learning approach (AS-MAML) by integrating attention mechanisms into residual networks for pedestrian gait recognition using frequency-modulated continuous-wave (FMCW) millimeter-wave radar through meta-learning. We enhance the feature extraction capability of the base network using channel attention mechanisms and integrate the additive angular margin loss function (ArcFace loss) into the inner loop of MAML to constrain inner loop optimization and improve radar discrimination. Then, this network is used to classify small-sample micro-Doppler images obtained from millimeter-wave radar as the data source for pose recognition. Experimental tests were conducted on pose estimation and image classification tasks. The results demonstrate significant detection and recognition performance, with an accuracy of 94.5%, accompanied by a 95% confidence interval. Additionally, on the open-source dataset DIAT-μRadHAR, which is specially processed to increase classification difficulty, the network achieves a classification accuracy of 85.9%. Full article

When considering a large dataset of histopathologic breast images captured at various magnification levels, the process of distinguishing between benign and malignant cancer from these images can be time-intensive. The automation of histopathological breast cancer image classification holds significant promise for expediting pathology diagnoses and reducing the analysis time. Convolutional neural networks (CNNs) have recently gained traction for their ability to more accurately classify histopathological breast cancer images. CNNs excel at extracting distinctive features that emphasize semantic information. However, traditional CNNs employing the softmax loss function often struggle to achieve the necessary discriminatory power for this task. To address this challenge, a set of angular margin-based softmax loss functions have emerged, including angular softmax (A-Softmax), large margin cosine loss (CosFace), and additive angular margin (ArcFace), each sharing a common objective: maximizing inter-class variation while minimizing intra-class variation. This study delves into these three loss functions and their potential to extract distinguishing features while expanding the decision boundary between classes. Rigorous experimentation on a well-established histopathological breast cancer image dataset, BreakHis, has been conducted. As per the results, it is evident that CosFace focuses on augmenting the differences between classes, while A-Softmax and ArcFace tend to emphasize augmenting within-class variations. These observations underscore the efficacy of margin penalties on angular softmax losses in enhancing feature discrimination within the embedding space. These loss functions consistently outperform softmax-based techniques, either by widening the gaps among classes or enhancing the compactness of individual classes. Full article

The present pilot study was designed by hypothesizing a possible correlation between lack of accuracy in implant placement and peri-implant hard and soft tissue health. A total of five patients underwent computer-guided implant surgery and full-arch immediate loading between 2013 and 2014. They subsequently underwent postoperative cone-beam computed tomography (CBCT). After a follow-up of 5 years, all patients were recalled for a clinical-radiographic evaluation of peri-implant health status. The mean linear deviation was 0.5 ± 0.2 mm at the implant’s head and 0.6 ± 0.2 mm at the implant’s apex, while the mean angular deviation of the long axis was 2.8° ± 1.2°. A mean marginal bone loss (MBL) of 1.16 ± 0.94 mm and 2.01 ± 1.76 mm was observed after 1 and 5 years of follow-up, respectively. At 5 years, the mean peri-implant probing depth (PPD) was 4.09 ± 1.44 mm, 66.6% of the evaluated implants showed peri-implant bleeding on probing (BOP), keratinized mucosa (KM) was <2 mm in 48.4% of cases, and mucosal recession (REC) ≥ 1 mm was assessed in 45.4% of the included implants. A negative correlation was observed between bucco-palatal/lingual linear inaccuracy and MBL, PPD, BOP, and KM. Full article

Traffic sign recognition is a pivotal technology in the advancement of autonomous vehicles as it is critical for adhering to country- or region-specific traffic regulations. Defined as an image classification problem in computer vision, traffic sign recognition is a technique that determines the class of a given traffic sign from input data processed by a neural network. Although image classification has been considered a relatively manageable task with the advent of neural networks, traffic sign classification presents its own unique set of challenges due to the similar visual features inherent in traffic signs. This can make designing a softmax-based classifier problematic. To address this challenge, this paper presents a novel traffic sign recognition model that employs angular margin loss. This model optimizes the necessary hyperparameters for the angular margin loss via Bayesian optimization, thereby maximizing the effectiveness of the loss and achieving a high level of classification performance. This paper showcases the impressive performance of the proposed method through experimental results on benchmark datasets for traffic sign classification. Full article