Search Results (1,080)

Background and Objectives: Faricimab, the first bispecific antibody targeting VEGF-A and angiopoietin-2, has demonstrated durable efficacy in pivotal phase 3 trials for neovascular age-related macular degeneration (nAMD) and diabetic macular edema (DME). Real-world data on treatment-naïve patients managed with fixed-interval maintenance protocols, particularly from South-Eastern Europe, remain limited. This pilot study evaluated 24-month outcomes of intravitreal faricimab in treatment-naïve nAMD and DME, using a standardized four-injection loading phase followed by fixed every-16-week (Q16W) maintenance. Materials and Methods: This study conducted a retrospective, observational, single-center pilot cohort study of 20 consecutive treatment-naïve eyes (9 nAMD, 11 DME). All patients received four monthly loading injections followed by a fixed every-16-week (Q16W) maintenance schedule, supplemented by discretionary additional injections for residual or recurrent disease activity (215 injections total; mean 10.75 ± 0.79 per patient; range 9–12). Primary outcomes were changes in central foveal thickness (CFT) and best-corrected visual acuity (BCVA; Snellen lines with ETDRS letter equivalents) at months 4 and 24. Prespecified secondary analyses included bootstrap 95% confidence intervals, a linear mixed-effects model with a time × disease-group interaction, Bayesian credible intervals with weakly informative priors, false-discovery-rate (FDR) correction, and a minimum detectable effect-size analysis. Results: All 20 eyes completed 24-month follow-up. In nAMD, mean CFT decreased by 186.9 ± 71.9 µm (35.9%; bootstrap 95% CI 148.1–236.0; p < 0.001; d = 2.60), and BCVA improved by 3.89 ± 0.78 Snellen lines (~19 ETDRS letters; 95% CI 3.44–4.33; p < 0.001; d = 4.97). In DME, CFT decreased by 197.7 ± 65.7 µm (39.3%; 95% CI 162.5–237.3; p < 0.001; d = 3.01), and BCVA improved by 4.55 ± 1.04 lines (~23 ETDRS letters; 95% CI 4.00–5.09; p < 0.001; d = 4.39). All 20 eyes (100%) achieved ≥ 3 Snellen lines gain and ≥20% CFT reduction; 80% reached final BCVA ≥ 7 lines. A linear mixed-effects model showed a significant time effect (p < 0.001) but no time × group interaction (CFT p = 0.84; BCVA p = 0.51), indicating concordant trajectories across diseases. Bayesian analysis with weakly informative priors yielded posterior P(|d| > 0.8) ≥ 0.99 for all primary outcomes. After FDR correction, all pre-specified primary comparisons remained significant. The minimum detectable effect size with the realized sample sizes (Cohen’s d ≈ 0.66 combined, 1.07 nAMD, 0.94 DME at 80% power) was substantially below all observed effect sizes. No ocular or systemic adverse events were recorded. Conclusions: In this small, single-center, treatment-naïve pilot cohort, a fixed Q16W faricimab maintenance schedule with discretionary additional injections was associated with durable anatomical and functional improvements over 24 months in both nAMD and DME, with no adverse events recorded across 215 injections. Given the limited sample, these findings should be regarded as hypothesis-generating. The high responder rates likely reflect the cohort’s substantial baseline visual impairment (mean baseline BCVA ~20/120–20/200), which provides greater absolute capacity for measurable gain than in higher-acuity registration trial populations. These pilot data support fixed-interval faricimab as a logistically feasible candidate strategy in resource-constrained settings and should be confirmed in larger multicenter cohorts using standardized ETDRS acuity assessment. Full article

Accurate beef cattle body measurement data are crucial for growth assessment, phenotypic analysis, breeding management, and precision livestock farming. Traditional manual measurements are labor-intensive, time-consuming, and likely to cause stress in animals, making it difficult to meet the demands of large-scale livestock farming. This paper employs a structured systematic literature review method, in accordance with the PRISMA 2020 guidelines, to summarize research progress in vision-based beef cattle body measurement. This paper focuses on reviewing technical approaches such as 2D image-based measurement, 3D measurement using RGB-D and LiDAR, and multi-view fusion. It analyzes key technologies including image segmentation, keypoint detection, point cloud processing, 3D reconstruction, and geometric calculations, and compares the advantages and disadvantages of different methods in terms of measurement accuracy, robustness, cost, and farm applicability. The results indicate that 2D image-based methods are low-cost and flexible to deploy but have limited expressiveness for 3D body measurement parameters; RGB-D and LiDAR methods can provide spatial information but are affected by point cloud noise, occlusion, equipment costs, and data processing complexity; multi-view fusion can improve the completeness of body surface information but places high demands on calibration, registration, and system integration. Current research still faces challenges such as a lack of public datasets, inconsistent annotation standards, uncertainty regarding ground truth, insufficient cross-ranch generalization validation, and limited practical applications. Future research should focus on developing standardized datasets, conducting cross-scenario validation, advancing multimodal perception, creating lightweight models, and applying edge computing to drive the evolution of visual body measurement toward continuous monitoring and intelligent decision-making. Full article

Accurate lens distortion correction is important for calibration, registration, image stitching, and 3D reconstruction, especially in low-data device-specific settings where disposable or specialised cameras cannot provide large calibration datasets. We address distortion correction for cameras with highly irregular or non-stationary distortion fields, where fixed polynomial models and generic learning-based rectification methods can struggle. We propose a framework based on Deep Gaussian Processes (DGPs) to model the non-linear mapping required for undistortion. The key motivation is that conventional single-layer GPs with stationary kernels must use one global notion of smoothness, whereas DGPs can represent spatially varying behaviour through composed latent mappings while preserving per-pixel predictive uncertainty. This uncertainty can be used to identify or downweight unreliable corrected regions in downstream tasks. We evaluate the method on three real camera datasets with increasing distortion complexity. The full structured acquisitions contain 512 horizontal and 512 vertical line images per camera. These are not thousands of natural calibration images, but they yield up to 29,532, 11,311, and 31,686 detected intersection correspondences for the RPI, Theta, and Pillcam datasets, respectively. This distinction is important for cameras where acquiring many independent images is impractical. The results are assessed using qualitative rectification, uncertainty maps, normalised collinearity errors, and total training time. Polynomial calibration remains strongest for the regular radial RPI distortion, while DGP and DGP2 models show lower normalised collinearity-error distributions than the standard GP and lightweight MLP baselines on the more distorted Theta and Pillcam datasets. For the full datasets, total DGP/DGP2 training times ranged from 2383.50 s to 10092.50 s, reflecting the additional computational cost of probabilistic non-stationary modelling. Full article

3D reconstruction of construction scenes is an important enabling technology for digital and intelligent construction project management. Recurring foreground occluders and dynamic disturbances in tower-crane imagery can destabilize image registration and introduce spurious depth responses. This paper proposes an occluder-mask-constrained 3D reconstruction framework driven by multi-view geometric anomalies. Adjacent-view geometric outliers are spatially aggregated to generate foreground prompt points, which are converted into occluder masks using Segment Anything Model 2 (SAM2). The masks are propagated as unified pixel-validity constraints through sparse feature filtering, Adaptive Patch Deformation Multi-View Stereo (APD-MVS) matching-cost evaluation, support-region selection, and depth-map fusion. Experiments on three real construction-site datasets show increased sparse-registration completeness in the tested sequences and fewer visually identifiable occluder-induced artifacts in dense point clouds. A representative 308-image sequence was further evaluated against no-mask reconstruction, You Only Look Once version 8 (YOLOv8) bounding-box removal, manually prompted Segment Anything Model 2.1 (SAM2.1), a Segment Anything Model 3 (SAM3) text-prompt baseline, and Visibility-Aware Multi-View Stereo Network (Vis-MVSNet). The evaluation combines sparse-reconstruction metrics, pixel-level mask-quality metrics from a manually annotated validation subset, module-wise runtime accounting, controlled ablations, and aligned dense-point-cloud visualization. These results show improved sparse-stage registration completeness and visible artifact suppression. Because high-precision 3D reference point clouds are unavailable, the dense results are interpreted as visual evidence of artifact suppression rather than as proof of improved absolute dense-reconstruction accuracy. Full article

In recent decades, humanoids have become more popular in various applications. However, their applications in human life are more than those in industry. In this paper, a humanoid is used to capture the sets of point clouds of an object for three-dimensional reconstruction. The structured light camera is widely used across diverse 3D scanning applications due to its high resolution, rapid acquisition capability, and adaptability to various material surfaces. Therefore, the humanoid developed by our team holds a structured light camera which captures the point clouds of an object put on a platform for the reconstruction of its 3D digital model. The platform is rotated so that the structured light camera can capture the image of all view angles on the object. Meanwhile, the structured light camera captures point clouds, and the camera of the humanoid recognizes the QR code on the platform so that the sets of point clouds can be distinguished by view angles on the object. Then, the automated registration process of the point cloud sets for a 3D model based on the point-to-plane iterative closest point (ICP) algorithm is proposed. The process incorporates preprocessing techniques, such as downsampling and normal vector estimated from plane, and utilizes the ICP algorithm for registration, ultimately achieving markerless and precision automatic merging of multi-view point cloud data. Experimental results demonstrate that the proposed method with the humanoid can effectively improve the completeness and accuracy of 3D reconstruction models, significantly reduce manual intervention, and enhance the system’s versatility and practical feasibility. Key parameters adjusted for more efficient computation of the ICP algorithm are revealed. In addition, the experimental results of the proposed ICP compared with G-ICP are also included. Full article

In precision livestock farming, 3D point clouds provide important data support for analyzing pig behavior and monitoring their health. However, due to environmental occlusions, limited sensor viewpoints, and mutual shielding between pigs, the acquired point clouds are often severely partial, which affects the accuracy of body shape modeling and behavior recognition. To address these challenges, this study constructed a pig pose point cloud dataset using multi-view depth camera acquisition and point cloud registration techniques. Based on this dataset, an improved point cloud completion model, IIR-PoinTr, is proposed to enhance the reconstruction of geometric and topological structures in pig bodies. By strengthening local geometric perception and high-dimensional feature representation, the model improves the reconstruction quality of partial pig point clouds and produces more structurally consistent pig body shapes. Experimental results show that, on the self-constructed pig posture dataset, the proposed method reduces Chamfer Distance (CD-L1) by 3.6%, CD-L2 by 6.9%, and Earth Mover’s Distance (EMD) by 2.0%, while improving the F-score by 5.4% compared with the baseline model. In single-view point cloud completion tasks, the method is capable of reconstructing geometrically consistent pig body structures and increases downstream classification accuracy by 34.9%. These results indicate that the proposed method can improve the reconstruction quality of partial pig point clouds and provide preliminary technical support for posture analysis under occlusion. Full article

Background/Objectives: Statistical shape modeling (SSM) is used to describe transtibial residual-limb morphology for prosthetic socket design, simulation, and future structural testing. However, conventional intrinsic metrics such as compactness, generality, and specificity may not directly reflect geometric fidelity to the original shape. This study examined the relationship between geometric fidelity and SSM evaluation and assessed a cross-sectional shape representation framework for transtibial residual limbs. Methods: Residual-limb surfaces were acquired from 62 adults with unilateral transtibial amputation using a structured-light 3D scanner while preserving habitual limb posture. Two surface-based registration methods, non-rigid iterative closest point and Bayesian coherent point drift, were compared with a cross-sectional representation in which proximal and distal regions were sectioned separately and reconstructed by strip triangulation. Geometric fidelity to the original mesh was quantified using average symmetric surface distance (ASSD). SSM performance was evaluated using compactness, generality, and specificity. Results: The optimal cross-sectional configuration was 60 sections × 72 points. The proposed method showed the best geometric fidelity (ASSD, 1.30 ± 0.14 mm), followed by Bayesian coherent point drift (1.33 ± 0.14 mm) and non-rigid iterative closest point (1.48 ± 0.48 mm). Compactness was highest for the proposed method, reaching 95% cumulative variance in four modes, compared with five and seven modes, respectively, for the two surface-based methods. In geometry-space evaluation, the proposed method showed the lowest specificity error, while differences in generality were statistically significant but small in magnitude. Conclusions: Intrinsic SSM metrics alone were insufficient to judge registration quality in transtibial residual-limb modeling. The cross-sectional representation preserved the original surface geometry more faithfully than the evaluated surface-based methods while maintaining competitive SSM performance. Full article

Extremely low overlap leads to severely scarce local geometric correspondences across frame pairs. Pure geometric descriptors—encoding merely low-level shape signatures—inherently fail to impose sufficient constraints for reliable transformation estimation when matches become critically sparse, rendering registration fundamentally fragile. While recent red-green-blue-depth (RGB-D) attempts have explored visual augmentation, they predominantly rely on low-level chromatic statistics or shallow convolutional neural network (CNN) features, underutilizing the rich hierarchical semantics inherent in RGB imagery. We present V3Reg, a robust registration framework that pioneers the integration of large-scale vision foundation models (DINOv3) with adaptive cross-modal fusion. Specifically, we extract mid-to-deep semantic features (Layer 11) from DINOv3 to transcend low-level texture limitations, and propose a Task-Aware Channel-Wise Gated Adaptive Fusion (TACGAF) module that dynamically calibrates geometric-visual contributions via registration-error-guided channel-wise gating. To rigorously evaluate ultra-low-overlap robustness, we reconstruct RGBD-ZeroMatch, a benchmark with controllable overlap ratios ranging from 1% to 20%. Extensive experiments demonstrate that V3Reg achieves 99.6% Feature Matching Recall and 96.3% Registration Recall on standard benchmarks. Notably, it maintains 50.2% Registration Recall at merely 5% overlap, outperforming prior methods by over 18 percentage points. Full article

Mechanical, electrical, and plumbing (MEP) pipe installation discrepancies can increase rework, complicate inspection, and affect subsequent operation and maintenance. This study presents a 3D laser scanning and Building Information Modeling (BIM)-based workflow for localizing and preliminarily classifying MEP pipe installation discrepancies in a building project. Preprocessed scanned pipe point clouds are registered with BIM-derived pipe point clouds through a coarse-to-fine Scan-BIM registration process. Individual pipe instances are extracted using distance-threshold-based growing, and scan-to-BIM pipe correspondence is established using nearest-neighbor root mean square error (RMSE). Pipes with relatively large overall RMSE values are further divided into slices to identify local high-discrepancy intervals. A slice-level discrepancy distribution function $R\left(s\right)$, together with derivative-magnitude and derivative-fluctuation thresholds, is used to support preliminary Type 1/Type 2 interpretation of representative discrepancy patterns. In a student dormitory case, the workflow screened local pipes with relatively large discrepancies, localized maximum-RMSE regions, and distinguished representative connection-related discrepancies from overall offset or inclination cases. A threshold perturbation check showed consistent Type 1/Type 2 labels for the four representative cases within the tested range. The workflow provides case-study evidence for localized MEP pipe inspection, while broader validation across projects and pipe systems remains necessary. Full article

Multi-view point cloud registration is a fundamental technology for 3D reconstruction and indoor robot navigation and remains a core challenge for robust environmental perception. Its key difficulty lies in achieving globally consistent alignment of multiple partially overlapping point clouds efficiently and reliably. To address the limitations of existing methods, including low registration accuracy under small overlaps, severe error accumulation in long sequences, and the difficulty of balancing computational efficiency with global consistency, this paper proposes a multi-view point cloud registration framework that integrates spanning tree-based global topology constraints with a multi-scale pyramid-based local refinement strategy, specifically validated for indoor environments. First, a Voxel-Guided Normal Consistency Keypoint Extraction (VG-NCKE) method is presented. It leverages voxel grids to guide stable computation of local geometric features and filters candidate keypoints using a neighborhood normal direction consistency metric, effectively improving keypoint repeatability and spatial uniformity on unevenly distributed point clouds. Second, a coarse registration strategy with global constraints is constructed based on the Overlap Confidence-weighted Minimum Spanning Tree (OC-WST). It quantifies inter-frame overlap reliability as edge weights and employs Prim’s algorithm to build the minimum spanning tree as the topological skeleton for global registration. By prioritizing high-overlap frame pairs, the method suppresses error propagation and reduces the complexity of multi-view registration. Additionally, a multi-scale pyramid ICP fine registration algorithm is designed. It adopts a point-to-plane error model instead of the traditional point-to-point distance metric and performs progressive optimization through a three-layer point cloud pyramid from coarse to fine. This expands the convergence basin and gradually improves alignment accuracy, mitigating the sensitivity of single-scale ICP to initial poses. Extensive experiments on the indoor 3DMatch dataset and real indoor LiDAR sequences demonstrate that the proposed method outperforms competing approaches in terms of registration accuracy, computational efficiency, and long-sequence robustness, validating its effectiveness for indoor multi-view point cloud registration tasks. Full article

Preserving cultural heritage sites requires an interoperable digital framework capable of integrating heterogeneous spatial data and supporting immersive interaction for inspection and management. This study investigates the integration of multiple heritage data representations—including IFC-based Heritage Building Information Modeling (HBIM), terrestrial and UAV LiDAR point clouds, and 3D Gaussian Splatting reconstructions—into a unified digital management environment for the East Hall (Aula Timur) heritage site within the Bandung Institute of Technology (ITB) campus. A semantic–spatial interoperability workflow is proposed to harmonize BIM, point cloud, and landscape-scale data within a common georeferenced context, supported by a CityGML-based base map of the surrounding site. An immersive virtual environment was implemented using a head-mounted display to enable walkthrough-based inspection and damage annotation. All datasets were georeferenced within a unified coordinate system, allowing spatial registration between digital objects and the physical heritage site. The results demonstrate that multi-source heritage datasets can be integrated with high geometric accuracy, achieving TLS registration errors of approximately 2 mm and georeferencing residuals within 11.1 cm (horizontal) and 0.95 cm (vertical), while preserving semantic information and ensuring spatial coherence across HBIM, GIS, and immersive environments. The system is implemented in VR, with an architecture designed to support future MR-based on-site annotation and visualization. The proposed framework establishes a foundation for future heritage digital twin deployments and supports informed conservation decisions. Full article

In UAV-based remote sensing, accurate and efficient image mosaicing is crucial for achieving real-time monitoring. Traditional cloud-centric processing paradigms, however, face core scientific challenges such as high latency, bandwidth bottlenecks, and limited autonomy, making them inadequate for dynamic, real-time scenarios. To address these issues, this paper proposes an edge-computing-enabled UAV image mosaicing system. The system consists of a UAV remote sensing platform and an edge computing terminal, with the core being our novel B-SIFT-ILS algorithm. The algorithm first uses geographic coordinates for unified registration, constructs a Gaussian scale space for multi-resolution representation, and then precisely locates extrema in the Difference of Gaussian (DoG) space using a 3D quadratic function. A BANSAC algorithm is subsequently employed to refine feature points and extract stable SIFT features, and finally, Iterative Least Squares (ILS) are used to achieve seamless mosaicing. Experimental results demonstrate that, compared with classical RANSAC, the proposed method achieves superior feature sampling accuracy (rotation: 0.879, translation: 0.877) and lower latency. The ILS-based smoothing stage effectively eliminates noise and ghosting without introducing gradient reversal, performing comparably to deep learning methods while significantly outperforming direct averaging and Gaussian approaches. On the NVIDIA Jetson Orin NX edge terminal, a single processing instance requires only 1124 ms, highlighting its strong potential for real-time, low-latency, and autonomous mosaicing tasks. Future research will focus on extending the approach to non-planar terrains and implementing adaptive parameter tuning for the BANSAC algorithm. Full article

Precise dimensional conformity of precast concrete members is critical for structural performance and on-site assembly accuracy, yet conventional manual inspection remains labor-intensive and unable to scale with modern production-line throughput. Existing scan-vs-BIM approaches address geometric verification in principle but are constrained by manual registration dependencies, the absence of machine-readable IFC-linked tolerance criteria, and limited validation under real factory yard conditions. This study presents a production-scale automated shape quality control (SQC) framework that closes all three gaps simultaneously. A purpose-designed two-point target device enables fully automated, repeatable registration seed-point extraction. A formal IFC property-set-linked rule engine architecture—comprising entity extraction, deviation computation, rule interpretation, and pass/fail decision stages—replaces ad hoc script-based tolerance checking with an interoperable, auditable compliance pipeline. Factory-scale validation on precast arch segments (n = 10) and wall panels (n = 12) achieved registration RMSE of 1.25–1.95 mm, pass rates exceeding 91%, and a 37.1% reduction in inspection time versus manual methods (95% CI: 34.5–39.6%; p < 0.001; Cohen’s d = 3.89). Repeatability testing yielded ICC = 0.971 and Bland–Altman limits of agreement of [−0.45, +1.07] mm. The framework represents a substantive step toward fully digital, production-integrated quality management for industrialized precast construction. Full article

The digital preservation of built cultural heritage requires precise documentation techniques capable of capturing complex architectural geometries often affected by occlusions and data voids. This study presents a robust multi-sensor fusion workflow integrating Terrestrial Laser Scanning (TLS) and Unmanned Aerial Vehicle (UAV) photogrammetry for the 3D reconstruction of the Hasaköy (Sasima) Church in Niğde, Türkiye. To address the limitations of traditional registration methods, specifically the susceptibility of the Iterative Closest Point (ICP) algorithm to local minima in datasets with partial overlaps, this study proposes a fine-tuning approach based on the Multi-population Based Differential Evolution (MDE) algorithm. The methodology employs a coarse-to-fine strategy, initiating with Fast Point Feature Histogram (FPFH) extraction and RANSAC (Random Sample Consensus) for global alignment, followed by TR-ICP, MDE, PSO, and Aquila Optimizer (AO) evaluation, computational-time analysis, FPFH-radius sensitivity testing, and 6-DoF transformation decomposition to characterize both accuracy and operational cost. In the 30-run fine-tuning evaluation, MDE reduced the mean bidirectional trimmed RMSE from 0.4152 m for TR-ICP to 0.3726 m. With a population parameter of 10, MDE retained a low median RMSE of 0.3718 m, while PSO exhibited a wider stochastic tail under the same bounded 6-DoF search budget. AO produced a higher mean bidirectional trimmed RMSE of 0.5233 m. The decimeter-scale bidirectional RMSE should be interpreted as a cross-source, partial-overlap distance metric rather than sensor precision; the overlapping facade objective was approximately 2.4–2.8 cm, and the UAV block was independently controlled with a 1.34 cm GCP RMSE. This study establishes a transparent and reproducible framework for heritage documentation, supporting the faithful digital preservation of endangered monuments with complex typologies. Full article

Single-view RGB-D observations are often affected by occlusion and restricted viewpoints, leading to incomplete object geometry and underestimated obstacle extents in indoor robot perception. This paper proposes a single-view scene completion framework that integrates candidate model retrieval and scale-aware registration. The framework first generates local RGB crops and partial point clouds through automatic instance segmentation; then retrieves complete candidate models by matching the local crops with multi-view rendered CAD images; and finally estimates candidate-to-observation rotation, translation, and scale to insert the selected aligned model into the original scene coordinate system. Experiments show that the retrieval module achieves Recall@1/Recall@5 of 80%/89%. The registration module reaches a success rate of 56.61%, outperforming the second-best method by 12.28 percentage points. More importantly, scene-level evaluation shows that the proposed method improves occupancy F1 from 0.445 to 0.523 and reduces boundary error from 0.202 m to 0.146 m compared with DiffCAD. These results indicate that the proposed framework improves navigation-oriented occupancy and obstacle-boundary recovery under CAD-library-based and segmentation-dependent single-view scene completion settings. Full article