Precise Reconstruction Research Articles

Signal Reconstruction Based on Time‐Varying Sliding Window in WSNs Using Compressed Sensing

ABSTRACTThis paper presents a new algorithm that utilizes compressed sensing (CS) for reconstruction of wireless sensor networks (WSNs) data with spatial and temporal correlation. The proposed method utilizes a time‐varying sliding window mechanism that dynamically adjusts both the window size and the number of measurements. This flexibility allows the algorithm to exploit spatio‐temporal correlations effectively, ensuring that data within the window remains sparse and thus more compressible. By dynamically varying the number of measurements, the algorithm equitably distributes the sampling rate across different time slots, adapting to changes in signal characteristics and minimizing transmission costs. Simulation results demonstrate that our proposed algorithm outperforms other CS reconstruction methods by achieving higher reconstruction precision while requiring fewer transmissions. This is achieved through a decentralized data‐window framework that maximizes the use of prior signal information, leading to improved signal recovery performance in diverse WSN scenarios.

Deep Neural Networks for Accurate Depth Estimation with Latent Space Features

Depth estimation plays a pivotal role in advancing human–robot interactions, especially in indoor environments where accurate 3D scene reconstruction is essential for tasks like navigation and object handling. Monocular depth estimation, which relies on a single RGB camera, offers a more affordable solution compared to traditional methods that use stereo cameras or LiDAR. However, despite recent progress, many monocular approaches struggle with accurately defining depth boundaries, leading to less precise reconstructions. In response to these challenges, this study introduces a novel depth estimation framework that leverages latent space features within a deep convolutional neural network to enhance the precision of monocular depth maps. The proposed model features dual encoder–decoder architecture, enabling both color-to-depth and depth-to-depth transformations. This structure allows for refined depth estimation through latent space encoding. To further improve the accuracy of depth boundaries and local features, a new loss function is introduced. This function combines latent loss with gradient loss, helping the model maintain the integrity of depth boundaries. The framework is thoroughly tested using the NYU Depth V2 dataset, where it sets a new benchmark, particularly excelling in complex indoor scenarios. The results clearly show that this approach effectively reduces depth ambiguities and blurring, making it a promising solution for applications in human–robot interaction and 3D scene reconstruction.

How can statistical models improve the accuracy of phylogenetic tree reconstruction?

When building the phylogenetic tree, current methods for calculating phylogenetic trees often lack the desired level of accuracy. This research paper explores the role of statistical models in enhancing the precision of phylogenetic tree reconstruction. The phylogenetic tree plays an important role in phylogenetic analysis and evolutionary biology. An accurate phylogenetic tree underpins our understanding of the major transitions in evolution, such as the emergence of new body plans or metabolism. Recent advancements in statistical modeling have more elaborate improvements than traditional methods. For example, Bayesian inference and maximum likelihood estimation provide frameworks for evaluating phylogenetic relationships by incorporating probability distributions and likelihood functions. These models account for the inherent uncertainties and variations in genetic data, leading to more accurate tree constructions. Moreover, using statistical models allows for the incorporation of complex evolutionary processes such as variable rates of evolution across lineages, horizontal gene transfer, and hybridization events. Techniques like Markov Chain Monte Carlo (MCMC) and bootstrap methods enhance the reliability of the inferred phylogenies by providing measures of confidence for the estimated relationships. Using these advanced statistical approaches, researchers can gain more accurate and reliable phylogenetic trees. Through a comprehensive review of current methodologies and case studies, this study aims to highlight statistical models' significant impact on evolutionary biology and the broader field of phylogenetics analysis and evolutionary biology.

Research on 3D reconstruction and digital protection of woodcarving works based on the combination of U-Net model and 6G network

In this study, we introduce an advanced method for the digital preservation of wood carvings, a significant component of our cultural heritage. By merging the U-Net model with 6G network technology, we’ve developed a precise and efficient 3D reconstruction process. The U-Net model achieved outstanding semantic segmentation with an average IoU of 0.97 and a Dice coefficient of 0.98, ensuring the detailed capture of wood carving features. The integration with 6G networks resulted in remarkably low data transmission delays—0.98 ms for transmission, 3.04 ms for processing, and 0.91 ms for queuing—highlighting the potential of 6G in supporting real-time, high-fidelity 3D reconstruction. This innovative approach not only meets the demands of digital preservation but also opens new avenues for protecting and promoting our cultural legacy, as affirmed by the high satisfaction scores from expert evaluations.

End to end polysemantic cooperative mixed task trainer for UAV target detection

With the rapid advancement and application of Unmanned Aerial Vehicles (UAVs), target detection in urban scenes has made significant progress. Achieving precise 3D reconstruction from oblique imagery is essential for accurate urban object detection in UAV images. However, challenges persist due to low detection accuracy caused by subtle target features, complex backgrounds, and the prevalence of small targets. To address these issues, we introduce the Polysemantic Cooperative Detection Transformer (Pc-DETR), a novel end-to-end UAV image target detection network. Our primary innovation, the Polysemantic Transformer (PoT) Backbone, enhances visual representation by leveraging contextual information to guide a dynamic attention matrix. This matrix, formed through convolutions, captures both static and dynamic features, resulting in superior detection. Additionally, we propose the Polysemantic Cooperative Mixed-Task Training scheme, which employs multiple auxiliary heads for diverse label assignments, boosting the encoder’s learning capacity. This approach customizes queries and optimizes training efficiency without increasing inference costs. Comparative experiments show that Pc-DETR achieves a 3% improvement in detection accuracy over the current state-of-the-art MFEFNet, setting a new benchmark in UAV image detection and advancing methodologies for intelligent UAV surveillance systems.

Application of data partitioned Kriging algorithm with GPU acceleration in real-time and refined reconstruction of three-dimensional radiation fields

Accurately assessing personal radiation doses in real radiation environments like nuclear power plants requires precise and real-time reconstruction of three-dimensional radiation fields. The Kriging algorithm, known for its accuracy in spatial interpolation, provides a promising approach for this task. However, its computational demands can be significant, especially in real-time scenarios. To address this, we enhance the Kriging algorithm with GPU acceleration and data partitioning strategies, enabling efficient and accurate reconstruction of three-dimensional nuclear radiation fields. Using Fluka software for Monte Carlo simulations, we generated a virtual radiation field of dimensions 5 m × 5 m × 5 m for a single-source, unshielded scenario, and a field of dimensions 20 m × 6 m × 8 m for a multi-source, shielded scenario. Using the simulated data, we compared the prediction accuracy of the improved algorithm with the conventional Kriging algorithm and further explored factors influencing the acceleration ratio of the improved algorithm. The results indicate that the GPU-accelerated and data-partitioned Kriging algorithm achieves nearly identical accuracy compared to the traditional method. In the single-source, unshielded scenario, with more than 343 known (measurement) points and predicting 95×95×95= 857,375 points, the prediction accuracy remains above 92.25%. In the multi-source, shielded scenario, with more than 8000 known (measurement) points and predicting 95×95×95= 857,375 points, the prediction accuracy remains above 91.17%. The acceleration performance of the improved algorithm is consistent across both scenarios, with the acceleration ratio increasing as the number of known and predicted points grows, reaching approximately 20 for smaller datasets and up to 93 for larger datasets. Additionally, the acceleration effect of the improved algorithm varies with data partition size, initially increasing and then decreasing as the partition size increases. When the number of known points is 512 and the number of predicted points is 884,736, the optimal partition size lies between 80,000 and 90,000, resulting in a prediction time of only 0.24 s.

Replicability of paleotemperature records in the northern Okinawa Trough and its implications for paleoceanographic reconstructions

Geochemical proxies are frequently utilized in the reconstruction of past ocean temperatures. Due to resource constraints, these reconstructions typically rely on a single sediment core, raising questions about the local and regional representativeness of paleotemperature records. To address this, we analyzed four sediment cores located within a 10-km radius in the northern Okinawa Trough (OT), which share the same climatic forcing and thus should reflect similar climate variations. We compiled published data and generated new paleotemperature estimates based on three widely used geochemical proxies (foraminiferal Mg/Ca, U37K′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{U}}_{37}^{{{\ ext{K}}^{\\prime}}}$$\\end{document}, TEX86\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{TEX}}_{86}$$\\end{document}). Analysis of the mean absolute deviations for nearby records based on the same proxy revealed that U37K′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{U}}_{37}^{{{\ ext{K}}^{\\prime}}}$$\\end{document} has the highest reproducibility, followed by Mg/Ca and TEX86\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{TEX}}_{86}$$\\end{document}. However, inconsistencies in inter-proxy offsets among nearby sites suggest the presence of noise in the proxy records, likely stemming from instrumental errors and sediment heterogeneity. Furthermore, the Mg/Ca and U37K′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{U}}_{37}^{{{\ ext{K}}^{\\prime}}}$$\\end{document} paleotemperature records agree within uncertainty when accounting for inter-site variability and calibration uncertainties, challenging previous interpretations of temperature signals from different seasons. All proxies indicate similar glacial-interglacial trends, albeit with varying magnitudes of temperature change. Both Mg/Ca and U37K′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{U}}_{37}^{{{\ ext{K}}^{\\prime}}}$$\\end{document} records suggest a glacial cooling of ~ 3 °C, whereas TEX86\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{TEX}}_{86}$$\\end{document} sea surface temperature (SST) data indicate a stronger glacial cooling of approximately ~ 6–8 °C. Modern observations indicate a subsurface TEX86\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{TEX}}_{86}$$\\end{document} recording depth of 50–100 m, coinciding with the thermocline. However, the TEX86\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{TEX}}_{86}$$\\end{document} subsurface temperature (subT) record does not resemble the Mg/Ca records of thermocline-dwelling foraminifera species. Instead, there is a better agreement with benthic foraminiferal Mg/Ca records of Uvigerina spp. (~ 700 m) and the intermediate temperature record derived from radiolarian assemblages (~ 500 m), pointing to a TEX86\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${\ ext{TEX}}_{86}$$\\end{document} recording depth that is deeper than the thermocline. In summary, our findings show that proxy noise can impact inter-proxy comparisons of paleotemperature records, but not the direction of glacial-interglacial shifts. Future research should prioritize constraining the recording depth of paleotemperature proxies and reducing calibration uncertainty for more precise and reliable quantitative paleotemperature reconstruction.

Binocular composite grayscale fringe projection profilometry based on deep learning for single-shot 3D measurements

The efficiency and precision of 3D shape reconstruction has been a focus in the fringe projection profilometry (FPP). However, achieving high-quality 3D measurement for isolated or overlapping objects from single fringe image is still a challenging task in the field. In this paper, a binocular composite grayscale fringe projection profilometry (BCGFPP) based on deep learning is proposed, in which a two-stage one-to-three network (TONet) is trained to predict the images required for phase unwrapping. The obtained absolute phase map exhibits high precision and eliminates deviation error and periodic ambiguity typically encountered in traditional sinusoidal composite fringe coding scheme. Haar transform principle is employed to form a Haar-like composite fringe pattern (HCFP), which consists of three different frequencies, serving as the input. TONet architecture is designed to predict images required by the tri-frequency four-step phase-shifting method (TFPM). Further, the absolute phase is calculated and the disparity map is obtained by matching the absolute phase of the left and right cameras. Finally, the 3D shape of the object can be restored by the system calibration parameters. Experimental results demonstrate the approach can greatly reduce the number of fringes required and achieve the accuracy of the absolute phase close to the training set.

Automated Approach to Accurate, Precise, and Fast Detector Simulation and Reconstruction.

Detector simulation and reconstruction are a significant computational bottleneck in particle physics. We develop particle-flow neural-assisted simulations (parnassus) to address this challenge. Our deep learning model takes as input a point cloud (particles impinging on a detector) and produces a point cloud (reconstructed particles). By combining detector simulations and reconstruction into one step, we aim to minimize resource utilization and enable fast surrogate models suitable for application both inside and outside large collaborations. We demonstrate this approach using a publicly available dataset of jets passed through the full simulation and reconstruction pipeline of the Compact Muon Solenoid (CMS) experiment. We show that parnassus accurately mimics the CMS particle flow algorithm on the (statistically) same events it was trained on and can generalize to jet momentum and type outside of the training distribution.

Research progress of three-dimensional printed customized prosthesis and its application in acetabular reconstruction of hip revision surgery

To review research progress on the design, manufacturing, and clinical application of three-dimensional (3D) printed customized prosthesis in acetabular reconstruction of hip revision surgery. The related research literature on 3D printed customized prosthesis and its application in acetabular reconstruction of hip revision surgery was searched by key words of "3D printed customized prosthesis", "revision hip arthroplasty", "acetabular bone defect", and "acetabular reconstruction" between January 2013 and May 2024 in Chinese and English databases, such as CNKI, Wanfang database, PubMed, etc. A total of 34 271 articles were included. After reading the literature titles, abstracts, or full texts, the literature of unrelated, repetitive, low-quality, and low evidence level was screened out, and a total of 48 articles were finally included for analysis and summary. The bone growth and mechanical properties of 3D printed customized prosthesis materials are better than those of non-3D printed customized prosthesis, which further solves the problem of elastic modulus mismatch between the implant and natural bone caused by "stress shielding"; the porous structure and antibacterial coating on the surface of 3D printed customized prosthesis have good anti-bacterial effect. 3D printed customized prosthesis can perfectly match the patient's individual acetabular anatomical characteristics and defect type, thus improving the accuracy of acetabular reconstruction and reducing the surgical time and trauma. 3D printed customized prosthesis can be used for precise and efficient individualized acetabular reconstruction in hip revision surgery with good early- and mid-term effectiveness. More optimized production technics and procedures need to be developed to improve the efficiency of clinical application and long-term effectiveness.

Study of shocks and ablation front in diamond ablator during a capsule implosion experiment at the National Ignition Facility

An X-ray phase contrast imaging platform using streaked refraction enhanced radiography (RER) was recently developed for capsule implosions at the National Ignition Facility. RER was demonstrated to image in-flight capsule density gradients such as the fuel-ablator interface that is not visible in traditional absorption only radiography. The latest experiments probing the early time evolution of the implosion allowed the precise measurement of the density gradients. An iterative analysis method has been applied to the RER radiograph to allow the reconstruction of temporal evolution of the radial density distribution from the ice-ablator interface to the ablation front. The estimated density reconstruction precision is ±2.4% with a density gradient sensitivity threshold of 1023cm−3 over a 2μm scale length. This enabled the study of shocks velocity and density gradients as well as ablation front scale length and shape.

Clinical outcomes after extra-articular resection of hip joint tumour using a custom-made osteotomy guide and 3D-printed endoprosthesis with posterior column preserved.

For rare cases when a tumour infiltrates into the hip joint, extra-articular resection is required to obtain a safe margin. Endoprosthetic reconstruction following tumour resection can effectively ensure local control and improve postoperative function. However, maximizing bone preservation without compromising surgical margin remains a challenge for surgeons due to the complexity of the procedure. The purpose of the current study was to report clinical outcomes of patients who underwent extra-articular resection of the hip joint using a custom-made osteotomy guide and 3D-printed endoprosthesis. We reviewed 15 patients over a five-year period (January 2017 to December 2022) who had undergone extra-articular resection of the hip joint due to malignant tumour using a custom-made osteotomy guide and 3D-printed endoprosthesis. Each of the 15 patients had a single lesion, with six originating from the acetabulum side and nine from the proximal femur. All patients had their posterior column preserved according to the surgical plan. Postoperative pathological assessment revealed a negative surgical margin was achieved in all patients. At final follow-up, 13.3% (2/15) died and no recurrence occurred. The overall survival was 81.7% at five years. None of the patients showed any signs of aseptic loosening, and no wound healing issues were observed. In total, 20% (3/15) developed complications, with two cases of early hip dislocation and one case of deep infection. The cumulative incidence of mechanical and non-mechanical failure in this series was 13.7% and 9.3%, respectively, at five years. In this cohort, the mean time to full weightbearing was 5.89 (SD 0.92) weeks and the mean Musculoskeletal Tumor Society score was 24.1 (SD 4.4). For patients with a hip joint tumour who met the inclusion criteria and were deemed suitable for posterior column preservation, a custom-made osteotomy guide combined with 3D-printed endoprosthesis is worth performing when treating patients who require extra-articular resection of the hip joint, as it can achieve adequate margin for local control, maximize bone preservation to maintain pelvic ring integrity, reduce the risk of complications by simplifying the surgical procedure, and allow for more precise reconstruction for better function.

Expanding Applications and Future of Robotic Microsurgery.

Robotic-assisted microsurgery has emerged as a transformative technology, offering enhanced precision for complex procedures across various fields, including lymphatic surgery, breast reconstruction, trauma, and neurosurgery. This paper reviews current advancements, applications, and potential future directions for robotic-assisted microsurgery. In lymphatic surgery, robotic systems such as Symani have improved precision in thoracic duct reconstruction and lymphatic vessel anastomoses, reducing morbidity despite longer surgery times. In breast reconstruction, robotic systems are being used to refine techniques like the miraDIEP approach, minimizing tissue damage and enhancing precision in individualized treatments. Trauma reconstruction, particularly for extremities, has also benefited from robotic assistance, enabling successful sutures in small vessels and nerves. Emerging applications in meningeal lymphatics show potential for treating neurodegenerative diseases through improved drainage. In neurosurgery, robots enhance precision in deep and narrow anatomic spaces, although advancements in specialized instruments are needed for full implementation. Future development of robotic microsurgery systems will focus on improved maneuverability, miniaturization, and integration of tools like augmented reality and haptic feedback. The goal is to combine robotic precision, data storage, and processing with human skills such as judgment and flexibility. Although robots are unlikely to replace surgeons, they are poised to play an increasingly significant role in enhancing surgical outcomes. As the technology evolves, further research and clinical trials are needed to refine robotic systems and validate their expanding applications in clinical practice.

A novel 3D uniformity measurement method in mechanical stirring systems

AbstractAccurate assessment of mixing uniformity is crucial in industrial mixing processes. This study proposes an evaluation method for three‐dimensional (3D) mixing processes that combines dual‐camera positioning and point pattern density fluctuation (PD) based on disordered hyperuniformity. This study employs a positioning method using dual‐camera to achieve precise capture and reconstruction of tracer particles in 3D space. The 3D reconstruction data is then evaluated for mixing performance using the PD method. A relationship model between |k| and time I values with mixing time was established for λ = 1. The results indicate that mixing time decreases with the increase of |k| and decreases with the decrease of I values. To ensure the accuracy of the PD method, feasibility analysis was conducted using conductometry. Additionally, the superiority of the PD method was validated by comparing it with the 3D‐Q method. The impact of bottom height of stirring paddle and motor speed on mixing effect were also investigated. This study establishes a fundamental groundwork and theoretical framework for optimizing parameters of stirring systems and assessing 3D mixing uniformity. It also offers important references and insights for engineering practices and theoretical research in related fields.

Impact of image-guided surgical techniques in complex bone reconstruction for maxillofacial fractures

Maxillofacial fractures present complex surgical challenges that require precise reconstruction techniques. Image-guided surgical methods, including computer navigation and 3D imaging, are increasingly used to improve surgical outcomes, yet selecting the optimal modality remains difficult due to the varying benefits of each technology. The databases used in the current research include, but are not limited to, PubMed, Scopus, and Google Scholar for article searches. Controlling for sources of bias, the review selected publications on image-guided surgical approaches to managing craniofacial fractures. Studies were identified based on the following criteria: methodology, relevance and patient-oriented results. Thus, analytical categories were developed to evaluate the results, including surgical meticulousness, complication incidences healing, and overall patient satisfaction. The research shows that the availability of enhanced imaging tools and navigation, including CT-nav, 3D-image ster, and intra-operative CBCT, enhances surgical precision and diagnostic accuracy. Preoperative planning repeatability when using CAN was up to 86.5%, and the need for revisions in maxillofacial trauma patients was reduced. Research on computation mirroring for navigation found differences between 0.12 mm, revealing enhanced surgery accuracy. Evaluations of image-guided techniques indicate their advantage over other methods in maxillofacial fracture operations in terms of accuracy and efficiency. Even nowadays, they provide significant improvements in patients’ treatment and surgical precision despite certain shortcomings in terms of technology and availability. Further studies need to consider enhancements to the navigation systems and consider cost benefits, depending on the type of setting in which the patient is located.

A deep learning framework for solving the prediction and reconstruction problem of Bingham fluid flow field

As a typical non-Newtonian fluid, Bingham fluid is employed in a multitude of fields, including petroleum, construction, and the chemical industry. However, due to the intricate intrinsic properties of Bingham fluids and the necessity for precision and efficacy in specific engineering applications, the rapid and precise prediction and reconstruction of its flow field information has become a challenge and a focal point of contemporary research. In this paper, we introduce a novel deep-learning approach to address the two-dimensional laminar motion of Bingham fluids. The proposed Papanastasiou Regularization Physics-Informed Neural Network (PR-PINN) framework effectively predicts and reconstructs the flow field of Bingham fluids. Initially, the framework applies Papanastasiou regularization to the governing equations of Bingham fluids, enhancing the network's adaptability to solving the flow field problem by incorporating boundary conditions and an adaptive weight assignment strategy. We consider two scenarios: equal-diameter circular pipe flow and conical pipe flow. The PR-PINN network is utilized for flow field prediction and reconstruction. Our results show that PR-PINN achieves high accuracy in flow field prediction and can reconstruct velocity and pressure fields using limited measurement data. Based on these findings, we explore the impact of boundary constraints, the effect of large intrinsic parameters on prediction accuracy, and the influence of measurement points and boundary constraints on flow field reconstruction. In summary, the PR-PINN network exhibits satisfactory performance and significant potential for predicting and reconstructing Bingham fluid flow fields.

Efficient inverse-kinematics solver for precise pose reconstruction of skinned 3D models

Efficient inverse-kinematics solver for precise pose reconstruction of skinned 3D models

RECONSTRUCTION OF THE SURFACE PROFILE OF OPTICAL-MECHANICAL STRUCTURE WITH WHITE LIGHT INTERFERENCE BY DEEP LEARNING U-NET MODEL

This research proposes a method to accurately determine the location of the interference signal peak in white light interferometry. This improves the accuracy and resolution of the measurement, enabling precise measurement and reconstruction of three-dimensional microstructures on optical surfaces. By using a deep learning algorithm with a U-net neural network architecture to generate a predicted signal that closely matches the reference signal, combined with Fast Fourier Transform to filter noise and fit the proposed function, the exact location of the envelope signal peak is determined, providing information about the height of the point. This method can achieve an accuracy of around 0.9 nanometers at a noise level of 50 dB, which is over 40% more accurate than the traditional Fourier transform method at various noise levels. The persistent issues of the traditional Fourier transform method, such as determining the location of the interference signal peak at a signal position and the accuracy being heavily influenced by noise and the fitting signal, are effectively addressed.

PV-LaP: Multi-sensor fusion for 3D Scene Understanding in intelligent transportation systems

Intelligent transportation systems are pivotal in modern urban development, aiming to enhance traffic management efficiency, safety, and sustainability. However, existing 3D Visual Scene Understanding methods often face challenges of robustness and high computational complexity in complex traffic environments. This paper proposes a Multi-Sensor Signal Fusion method based on PV-RCNN and LapDepth (PV-LaP) to improve 3D Visual Scene Understanding. By integrating camera and LiDAR data, the PV-LaP method enhances environmental perception accuracy. Evaluated on the KITTI and WHU-TLS datasets, the PV-LaP framework demonstrated superior performance. On the KITTI dataset, our method achieved an Absolute Relative Error (Abs Rel) of 0.079 and a Root Mean Squared Error (RMSE) of 3.014, outperforming state-of-the-art methods. On the WHU-TLS dataset, the method improved 3D reconstruction precision with a PSNR of 19.15 and an LPIPS of 0.299. Despite its high computational demands, PV-LaP offers significant improvements in accuracy and robustness, providing valuable insights for the future development of intelligent transportation systems.

Curved Image Triangulation Based on Differentiable Rendering

AbstractImage triangulation methods, which decompose an image into a series of triangles, are fundamental in artistic creation and image processing. This paper introduces a novel framework that integrates cubic Bézier curves into image triangulation, enabling the precise reconstruction of curved image features. Our developed framework constructs a well‐structured curved triangle mesh, effectively preventing overlaps between curves. A refined energy function, grounded in differentiable rendering, establishes a direct link between mesh geometry and rendering effects and is instrumental in guiding the curved mesh generation. Additionally, we derive an explicit gradient formula with respect to mesh parameters, facilitating the adaptive and efficient optimization of these parameters to fully leverage the capabilities of cubic Bézier curves. Through experimental and comparative analyses with state‐of‐the‐art methods, our approach demonstrates a significant enhancement in both numerical accuracy and visual quality.