Search Results (833)

The alarming realism of Deepfake presents a significant challenge to digital authenticity, yet its inherent difficulty in synchronizing the emotional cues between facial expressions and speech offers a critical opportunity for detection. However, most existing approaches rely on general-purpose backbones for unimodal feature extraction, resulting in an inadequate representation of fine-grained dynamic emotional expressions. Although a limited number of studies have explored cross-modal emotional consistency of deepfake detection, they typically employ shallow fusion techniques which limit latent expressiveness. To address this, we propose ACE-Net, a novel framework that identifies forgeries via multimodal emotional inconsistency. For the speech modality, we design a bidirectional cross-attention mechanism to fuse acoustic features from a lightweight CNN-based model with textual features, yielding a representation highly sensitive to fine-grained emotional dynamics. For the visual modality, a MobileNetV3-based perception head is proposed to adaptively select keyframes, yielding a representation focused on the most emotionally salient moments. For multimodal emotional consistency discrimination, we develop a multi-dimensional fusion strategy to deeply integrate high-level emotional features from different modalities within a unified latent space. For unimodal emotion recognition, both the audio and visual branches outperform baseline models on the CREMA-D dataset. Building on this, the complete ACE-Net model achieves a state-of-the-art AUC of 0.921 on the challenging DFDC benchmark. Full article

As a fundamental low-level vision task, image restoration plays a pivotal role in reconstructing authentic visual information from corrupted inputs, directly impacting the performance of downstream high-level vision systems. Current approaches frequently exhibit two critical limitations: (1) Progressive texture degradation and blurring during iterative refinement, particularly in irregular damage patterns. (2) Structural incoherence when handling cross-domain artifacts. To address these challenges, we present a semantic-aware hierarchical network (SAHN) that synergistically integrates multi-scale semantic guidance with structural consistency constraints. Firstly, we construct a Dual-Stream Feature Extractor. Based on a modified U-Net backbone with dilated residual blocks, this skip-connected encoder–decoder module simultaneously captures hierarchical semantic contexts and fine-grained texture details. Secondly, we propose the semantic prior mapper by establishing spatial–semantic correspondences between damaged areas and multi-scale features through predefined semantic prototypes through adaptive attention pooling. Additionally, we construct a multi-scale fusion generator, by employing cascaded association blocks with structural similarity constraints. This unit progressively aggregates features from different semantic levels using deformable convolution kernels, effectively bridging the gap between global structure and local texture reconstruction. Compared to existing methods, our algorithm attains the highest overall PSNR of 34.99 with the best visual authenticity (with the lowest FID of 11.56). Comprehensive evaluations of three datasets demonstrate its leading performance in restoring visual realism. Full article

Cooperative path planning is recognized as a critical technology for Autonomous Underwater Vehicle (AUV) clusters to execute complex marine operations. Through multi-AUV cooperative decision-making, perception limitations of individual robots can be mitigated, thereby significantly enhancing the efficiency of tasks such as deep-sea resource exploration and submarine infrastructure maintenance. However, the underwater environment is characterized by severe disturbances and limited communication, making cooperative path planning for AUV clusters particularly challenging. Currently, this field is still in its early research stage, and there exists an urgent need for the integration of scattered technical achievements to provide theoretical references and directional guidance for relevant researchers. Based on representative studies published in recent years, this paper provides a review of the research progress in three major technical domains: heuristic optimization, reinforcement and deep learning, and graph neural networks integrated with distributed control. The advantages and limitations of different technical approaches are elucidated. In addition to cooperative path planning algorithms, the evolutionary logic and applicable scenarios of each technical school are analyzed. Furthermore, the lack of realism in algorithm training environments has been recognized as a major bottleneck in cooperative path planning for AUV clusters, which significantly limits the transferability of algorithms from simulation-based validation to real-sea applications. This paper aims to comprehensively outline the current research status and development context of the field of AUV cluster cooperative path planning and propose potential future research directions. Full article

High-quality 3D objects play a crucial role in digital twins, while synthetic data generated from these objects have become essential in deep learning-based computer vision applications. The task of collecting and labeling real defects on industrial object surfaces has many challenges and efforts, while synthetic data generation feasibly replicates huge amounts of labeled data. However, synthetic datasets lack realism in their rendered images. To overcome this issue, this paper introduces a single framework for 3D industrial object reconstruction and synthetic defect generation for digital twin smart factory applications. In detail, NeRF is applied to reconstruct our custom industrial 3D objects through videos collected by a smartphone camera. Several NeRF-based models (i.e., Instant-NGP, Nerfacto, Volinga, and Tensorf) are compared to choose the best outcome for the next step of defect generation on the 3D object surface. To be fairly evaluated, we train four models using the Nerfstudio framework with our three custom datasets of two objects. From the experiment’s results, Instant-NGP and Nerfacto achieve the best outcomes, outperforming all other methods significantly. The exported meshes of 3D objects are refined using Blender before loading into NVIDIA Omniverse Code to generate defects on the surface with the Replicator. To evaluate the object detection performance and to verify the benefits of synthetic defect data, we conducted experiments with YOLO-based models on our synthetic and real-plus-synthetic defects. From the experiment’s results, the synthetic defect data contribute to improving YOLO models’ generalization capability with the highest and lowest accuracy mAP@0.5 enhancement of 18.8 and 1.5 percent on YOLOv6n and YOLOv8s, respectively. Full article

Flooding poses a major hazard to rapidly urbanising cities in Southeast Asia, and risks are projected to intensify under climate change. Accurate risk assessment, however, is hindered by scarcity of hydrological and topographic data. Focusing on the Lower Prek Thnot River Basin, a peri-urban catchment of Phnom Penh, Cambodia, the study applied the Rainfall–Runoff–Inundation model and systematically augmented inputs: hourly satellite rainfall data, field-surveyed river cross-sections and representation of hydraulic infrastructure such as weirs and pumping. Validation used Sentinel-1 SAR-derived flood-extent maps for the October 2020 event. Scenario comparison shows that rainfall input and channel geometry act synergistically: omitting either degrades performance and spatial realism. The best configuration (Sim. 5) Accuracy = 0.891, Hit Ratio = 0.546 and True Ratio = 0.701 against Sentinel-1, and reproduced inundation upstream of weirs while reducing overestimation in urban districts through pumping emulation. At the study’s 500 m grid, updating land use from 2002 to 2020 had only a minor effect relative to rainfall, geometry and infrastructure. The results demonstrate that targeted data augmentation—combining satellite products, field surveys and operational infrastructure—can deliver robust inundation maps under data scarcity, supporting hazard mapping and resilience-oriented flood management in rapidly urbanising basins. Full article

This study investigates the effectiveness of a virtual reality (VR) simulation for teaching the hand lay-up process in composite manufacturing within mechanical engineering education. A within-subjects experiment involving 17 undergraduate mechanical engineering students compared the VR-based training with conventional physical laboratory instruction. Task performance, cognitive load, and learner perceptions were measured using procedural accuracy scores, completion times, NASA-TLX workload ratings, and post-task interviews. Results indicated that while participants required more time to complete the task in VR, procedural accuracy was comparable between VR and physical labs. VR significantly reduced mental, physical, and effort-related demands but elicited higher frustration levels, primarily due to navigation challenges and motion discomfort. Qualitative feedback showed strong learner preference for VR, citing its hazard-free environment, repeatability, and step-by-step guidance. These findings suggest that VR offers a viable and pedagogically effective alternative or complement to traditional composite-manufacturing training, particularly in contexts where access to physical facilities is limited. Future work should examine long-term skill retention, incorporate haptic feedback for tactile realism, and explore hybrid models combining VR and physical practice to optimise learning outcomes. Full article

Large language model (LLM)-based virtual patient (VP) simulations are emerging to complement traditional medical and nursing education by enabling safe, repeatable, and context-rich clinical practice. This review synthesizes recent developments from 2023 to 2025, mapping implementation approaches, data practices, evaluation methods, and cross-cutting challenges across forty studies. Six implementation categories are identified: scenario generation; prompt-driven VPs; feedback-integrated automated scoring; realism- and adaptability-enhanced systems; knowledge-driven and multi-agent hybrids; and mental health-oriented systems. The analysis summarizes dataset usage (including knowledge sources and governance) and evaluation frameworks, and it introduces quantitative indicators for reproducible assessment. Persistent challenges include factual accuracy, role consistency, emotional realism, and ethical and legal accountability. Overall, LLM-based VP systems show growing potential to extend simulation-based learning, but stronger evidence from multi-site controlled studies, standardized metrics, transparent reporting (model versions, prompts), and robust data governance is needed to establish educational validity and generalizability. Full article

The integration of smartphone-powered Virtual Reality (VR) into architectural practice is transforming how unbuilt spaces are perceived. The presented study is based on ten single-family house projects in which immersive visualization was introduced through mobile VR headsets connected to Building Information Modeling (BIM). It should be treated as a pilot study, preceding further comprehensive research on the subject. A total of 23 participants (investors and future users of the buildings at the same time) were actively involved in the design process supported by VR technology. Field of view adjustment was implemented within the BIM + VR model to align the virtual perception with the natural human visual range, improving the realism of the experience. Preliminary findings indicated that VR walkthroughs enhanced the future users’ understanding of spatial arrangements and supported informed decision-making. Over 80% of participants reported that it helped them better assess room sizes, placement of windows and doors, and furniture layout. This improved communication between investors and designers, and reduced the number of revisions required at further design stages. The use of VR to merge architecture with interior design enabled a human-scale perspective, cost optimization, and the exploitation of BIM + VR visualization potential for sustainable residential design. Full article

The Discrete Element Method (DEM) has become a cornerstone for analysing granular flow and mixing phenomena in static mixers. This review provides a comprehensive synthesis that distinguishes it from previous studies by: (i) covering a broad range of static mixer geometries, including Kenics, SMX, and Sulzer designs; (ii) integrating experimental validation methods, such as particle tracking, high-speed imaging, Particle Image Velocimetry (PIV), and X-ray tomography, to assess DEM predictions; and (iii) systematically analyzing computational strategies, including advanced contact models, hybrid DEM-CFD/FEM frameworks, machine learning surrogates, and GPU-accelerated simulations. Recent advances in contact mechanics—such as improved cohesion, rolling resistance, and nonspherical particle modelling—have enhanced simulation realism, while adaptive time-stepping and coarse-graining improve computational efficiency. DEM studies have revealed several non-obvious relationships between mixer geometry and particle dynamics. Variations in blade pitch, helix angle, and element arrangement significantly affect local velocity fields, mixing uniformity, and energy dissipation. Alternating left–right element orientations promote cross-sectional particle exchange and reduce stagnant regions, whereas higher pitch angles enhance axial transport but can weaken radial mixing. Particle–wall friction and surface roughness strongly govern shear layer formation and segregation intensity, demonstrating the need for geometry-specific optimization. Comparative analyses elucidate how particle–wall interactions and channel structure influence segregation, residence time, and energy dissipation. The review also identifies current limitations, highlights validation and scale-up challenges, and outlines key directions for developing faster, more physically grounded DEM models, providing practical guidance for industrial mixer design and optimization. Full article

MyDispense is a high-fidelity, low-stakes community pharmacy simulation, allowing students to practice dispensing skills. A systematic review was conducted to identify students’ perceptions regarding barriers and facilitators of MyDispense in pharmacy education. PubMed, CINAHL, and EMBASE databases were searched from 2015 to 2025 in January 2025 using combined keywords, proximity searching and Boolean operators. Studies investigating MyDispense and gathering students’ perceptions were included. Record screening was conducted by two independent reviewers (OC and LS). Any identified records from database searching and hand searching of included study reference lists were imported to Rayyan and subjected to independent review. Conflicts were resolved through a third party (RMcC), and discussions were held until consensus was reached. Fifteen studies were included in this review. Seven studies were conducted in USA, six in Asia, one in UK, and one in Australia. All studies utilised purposive sampling. Sample sizes ranged from 33 to 322 students. All studies included surveys to gather student perceptions. Other data collection methods included semi-structured interviews and focus group discussions for students to further elaborate on survey responses. Identified facilitators were mapped to four overarching themes; “Develops competency”, “User-Friendliness”, “Engaging Learning Experience” and “Safe Learning Environment.” Key barriers were encompassed to three themes: “Learning Curve”, “IT issues” and “Limited Realism and Applications”. Barriers included (i) the learning curve of the platform, (ii) technical issues, and (iii) limited realism. Facilitators included perceptions of (i) improved dispensing and counselling skills and a deeper understanding of pharmacy legislation, (ii) accessibility, interactivity of the learning environment and (iii) immediate feedback. Synthesis of the evidence in this review identified students’ perceptions of barriers and facilitators of MyDispense in pharmacy education. This may serve as a guide to educators considering the adoption of MyDispense into their curricula. Full article

The mathematical universe of the quantum topos, which is formulated on the basis of classical Boolean snapshots, delivers a neo-realist description of quantum mechanics that preserves realism. The main contribution of this article is developing formal objectivity in physical theories beyond quantum mechanics in the topos-theory approach. It will be shown that neo-realist responses to non-perturbative structures of quantum field theory do not preserve realism. In this regard, the method of Feynman graphons is applied to reframe the task of describing objectivity in quantum field theory in terms of replacing the standard Hilbert-space/operator-algebra ontology with a new context category built from a certain family of topological Hopf subalgebras of the topological Hopf algebra of renormalization as algebraic/combinatorial data tied to non-perturbative structures. This topological-Hopf-algebra ontology, which is independent of instrumentalist probabilities, enables us to reconstruct gauge field theories on the basis of the mathematical universe of the non-perturbative topos. The non-Boolean logic of the non-perturbative topos cannot be recovered by classical Boolean snapshots, which is in contrast to the quantum-topos reformulation of quantum mechanics. The article formulates a universal version of the non-perturbative topos to show that quantum field theory is a globally and locally neo-realist theory which can be reconstructed independent of the standard Hilbert-space/operator-algebra ontology. Formal objectivity of the universal non-perturbative topos offers a new route to build objective semantics for non-perturbative structures. Full article

Traffic safety continues to be a pressing worldwide issue, with young drivers especially exposed to accidents because of limited experience, reckless behaviors, and risky practices such as driving under the influence of alcohol or other substances. In this scenario, reliable methods to evaluate psychomotor and sensory abilities essential for safe driving are highly needed. This study presents the development of a Virtual Reality (VR) prototype aimed at enhancing psychometric testing. The platform incorporates immersive environments to assess peripheral vision, reaction time, and motor accuracy, implemented with Oculus Quest 2, Blender, and Unity. The VR-based system was validated through black-box testing and user satisfaction surveys with a sample of 80 licensed drivers in single-session evaluations. The findings demonstrate that VR increases both precision and realism in psychomotor evaluations: 81.25% of participants perceived the scenarios as realistic, and 85% agreed that the system effectively measured critical driving skills. While a few users experienced minor discomfort, 97.5% recommended its application in practical assessments. This study highlights VR as a robust alternative to conventional psychometric/psychotechnical tests, capable of improving measurement reliability and user engagement and paving the way for more efficient and inclusive driver training initiatives. Full article

The monitoring of turbidity in estuarine environments is a challenging essential task for managing water quality and ecosystem health. This study focuses on the lower reaches of the River Mersey, Liverpool. Harmonized Sentinel-2 MSI Level-2A imagery was integrated with in situ measurements from seven Environment Agency monitoring stations for two consecutive years (January 2023–January 2025). The workflow included image preprocessing, spectral index calculation, and the application of four machine learning algorithms: Gradient Boosting Regressor, XGBoost, Support Vector Regressor, and K-Nearest Neighbors. Among these, Gradient Boosting Regressor achieved the highest predictive accuracy (R² = 0.84; RMSE = 15.0 FTU), demonstrating the suitability of ensemble tree-based methods for capturing non-linear interactions between spectral indices and water quality parameters. Residual analysis revealed systematic errors linked to tidal cycles, depth variation, and salinity-driven stratification, underscoring the limitations of purely data-driven approaches. The novelty of this study lies in demonstrating the feasibility and proof-of-concept of using machine learning to derive spatially explicit turbidity estimates under data-limited estuarine conditions. These results open opportunities for future integration with Computational Fluid Dynamics models to enhance temporal forecasting and physical realism in estuarine monitoring systems. The proposed methodology contributes to sustainable coastal management, pollution monitoring, and climate resilience, while offering a transferable framework for other estuaries worldwide. Full article

Flood risk in small streams is rising under climate change, as small catchments are highly vulnerable to short, intense storms. We develop a high-resolution assessment that integrates a Digital Surface Model (DSM), a Digital Elevation Model (DEM), and airborne LiDAR within a MATLAB (2025b) hydraulic workflow. A hybrid elevation model uses the DEM as baseline and selectively retains DSM-derived structures (levees, bridges, embankments), while filtering vegetation via DSM–DEM differencing with a 1.0 m threshold and a 2-pixel kernel. We simulate 10-, 30-, 50-, 100-, and 200-year return periods and calibrate the 200-year case to the July 2025 Sancheong event (793.5 mm over 105 h; peak 100 mm h⁻¹). The hybrid approach improves predictions over DEM-only runs, capturing localized depth increases of 1.5–2.0 m behind embankments and reducing false positives in vegetated areas by 12–18% relative to raw DSM use. Multi-frequency maps show progressive expansion of inundation; in the 100-year scenario, 68% of the inundated area exceeds 2.0 m depth, while 0–1.0 m zones comprise only 13% of the footprint. Unlike previous DSM–DEM studies, this work introduces a selective integration approach that distinguishes structural and vegetative features to improve the physical realism of small-stream flood modeling. This transferable framework supports climate adaptation, emergency response planning, and sustainable watershed management in small-stream basins. Full article

Fiber-reinforced polymer (FRP) laminates are widely used in both aviation and structural engineering, yet their implementation reflects fundamentally different paradigms. Aviation represents a fatigue-critical, certification-driven domain, while structural engineering emphasizes long-term durability and environmental resilience. These sectors were selected as conceptual extremes to explore how contrasting design philosophies, degradation mechanisms, and inspection strategies shape the performance and reliability of laminated FRP composites. Their approaches offer complementary insights: aviation contributes high-fidelity modeling and embedded monitoring, while structural engineering provides scalable inspection strategies and exposure-based degradation logic. Both sectors employ classical laminate theory and finite element modeling, but diverge in modeling depth and regulatory integration. This review synthesizes these contrasts based on 168 literature references, including 141 published between 2020 and 2025, reflecting recent developments in composite design, modeling, and inspection. It contributes to materials engineering by proposing hybrid modeling and inspection frameworks that integrate progressive damage simulation with durability-based design logic. By bridging the modeling precision of aviation with the environmental realism of structural engineering, this review outlines a pathway toward unified, sustainable, and adaptive engineering practices for laminated FRP composites. Full article