Improve Reconstruction Accuracy Research Articles

Neural Surfel Reconstruction: Addressing Loop Closure Challenges in Large-Scale 3D Neural Scene Mapping

Efficiently reconstructing complex and intricate surfaces at scale remains a significant challenge in 3D surface reconstruction. Recently, implicit neural representations have become a popular topic in 3D surface reconstruction. However, how to handle loop closure and bundle adjustment is a tricky problem for neural methods, because they learn the neural parameters globally. We present an algorithm that leverages the concept of surfels and expands relevant definitions to address such challenges. By integrating neural descriptors with surfels and framing surfel association as a deformation graph optimization problem, our method is able to effectively perform loop closure detection and loop correction in challenging scenarios. Furthermore, the surfel-level representation simplifies the complexity of 3D neural reconstruction. Meanwhile, the binding of neural descriptors to corresponding surfels produces a dense volumetric signed distance function (SDF), enabling the mesh reconstruction. Our approach demonstrates a significant improvement in reconstruction accuracy, reducing the average error by 16.9% compared to previous methods, while also generating modeling files that are up to 90% smaller than those produced by traditional methods.

A deep learning-based spatial gradient reconstruction method for efficient damage identification in composite with high-sparsity Lamb wavefield

The structural integrity and safety of carbon fiber reinforced plastics (CFRP) are vulnerable to delamination, which is often imperceptible to the naked eye. Although the Scanning Laser Doppler Vibrometer (SLDV) has shown promise in damage quantification of CFRP, its time-consuming measurement process limits its application in engineering scenarios. To address this, we introduce a novel damage index, the spatial gradient, which captures the interaction between delamination and the wavefield. We have also developed a neural network capable of reconstructing the spatial gradient directly from high-sparsity Lamb wavefield data obtained at an extremely low spatial sampling rate, thereby significantly reducing measurement time. To enhance the network’s capability to detect wavefield anomalies, we employ the cross-attention technique, allowing for the direct injection of shallow features representing local wavefield distortions caused by damage into the decoder. Additionally, we integrate multiple reconstruction layers to guide the wavefield reconstruction process, ensuring meaningful information is captured at each stage. Our method achieves substantial improvements in reconstruction accuracy, increasing from 70 % to 92 % in single-damage scenario and from 14 % to 72 % in multi-damage scenario compared to the previous state-of-the-art techniques. By using the reconstructed spatial gradient field for damage imaging through spatial covariance analysis, our approach demonstrates its feasibility and generalizability across various damage locations. This suggests its potential as a reliable solution for fast and accurate damage characterization, reducing the measurement burden and enhancing practical applicability.

RBP-DIP: Residual back projection with deep image prior for ill-posed CT reconstruction

The success of deep image prior (DIP) in a number of image processing tasks has motivated their application in image reconstruction problems in computed tomography (CT). In this paper, we introduce a residual back projection technique (RBP) that improves the performance of deep image prior framework in iterative CT reconstruction, especially when the reconstruction problem is highly ill-posed. The RBP-DIP framework uses an untrained U-net in conjunction with a novel residual back projection connection to minimize the objective function while improving reconstruction accuracy. In each iteration, the weights of the untrained U-net are optimized, and the output of the U-net in the current iteration is used to update the input of the U-net in the next iteration through the proposed RBP connection. The introduction of the RBP connection strengthens the regularization effects of the DIP framework in the context of iterative CT reconstruction leading to improvements in accuracy. Our experiments demonstrate that the RBP-DIP framework offers improvements over other state-of-the-art conventional IR methods, as well as pre-trained and untrained models with similar network structures under multiple conditions. These improvements are particularly significant in the few-view and limited-angle CT reconstructions, where the corresponding inverse problems are highly ill-posed and the training data is limited. Furthermore, RBP-DIP has the potential for further improvement. Most existing IR algorithms, pre-trained models, and enhancements applicable to the original DIP algorithm can also be integrated into the RBP-DIP framework.

Prior-image-based low-dose CT reconstruction for adaptive radiation therapy.

The study aims to reduce the imaging radiation dose in Adaptive Radiotherapy (ART) while maintaining high-quality CT images, critical for effective treatment planning and monitoring. We developed the Prior-aware Learned Primal-Dual Network (pLPD-UNet), which uses prior CT images to enhance reconstructions from low-dose scans. The network was separately trained on thorax and abdomen datasets to accommodate the unique imaging requirements of each anatomical region. The pLPD-UNet demonstrated improved reconstruction accuracy and robustness in handling sparse data compared to traditional methods. It effectively maintained image quality essential for precise organ delineation and dose calculation, while achieving a significant reduction in radiation exposure. This method offers a significant advancement in the practice of ART by integrating prior imaging data, potentially setting a new standard for balancing radiation safety with the need for high-resolution imaging in cancer treatment planning.

TC-SRNet: Reconstruction and prediction of typhoon high-resolution turbulent fields based on meteorological numerical forecast scale wind fields and deep learning method

Meteorological numerical forecast models can provide a more accurate typhoon wind fields compared to engineering typhoon models due to their incorporate atmospheric multiphysical processes. However, the latest advancements in supercomputing power indicate that the current finest level of real-time weather forecasting typically operates on grid scales ranging from 1 to 4 km, while the convergence of typhoon intensity and turbulent field characteristics occurs at scales as fine as 62–185 m. Therefore, the primary scientific inquiry lies in determining how to achieve high-precision turbulent wind fields while considering the realistic atmospheric multiphysical processes, that means establish a “bridge” of typhoon wind field between kilometer-level and hundred-meter scales. This study investigates the super-resolution reconstruction of wind fields across different horizontal grid scales, utilizing a benchmark wind field at a 62 m horizontal grid scale (ground truth), which is based on a hybrid down-sampling skip connection (DSC)/multi-scale (MS) model. The research findings demonstrate that compared to traditional interpolation methods, the DSC/MS method significantly improves reconstruction accuracy, albeit with some residual high-frequency energy dissipation issues. Additionally, the DSC/MS method currently exhibits better reconstruction performance for 62 m scale wind fields based on kilometer-scale and smaller horizontal grid scales (1 km, 555 m, 185 m), with improved reconstruction as grid scale decreases. However, significant errors are observed in reconstructing fine turbulent fields at 62 m scale based on wind fields at 1.67 km horizontal grid scale. The findings presented in the present study can provide real and high-precision turbulent wind fields for structural wind engineering and wind energy assessment studies, thereby holding significant scientific and engineering application value.

Study on the impact of camera ISP algorithms on stripe structured light 3D reconstruction and the improvement of 3D reconstruction accuracy by photon input

This paper investigates the impact of camera image signal processing (ISP) algorithms on stripe structured light 3D reconstruction and explores the improvement of 3D reconstruction accuracy by photon input. Existing 3D reconstruction technologies have broad applications in fields such as intelligent manufacturing, healthcare, and consumer electronics, but their high precision requirements often cannot be met by current ISP algorithms. By using handheld devices to capture images and combining key techniques such as the multi-step phase shifting method, multi-frequency phase unwrapping method, and triangulation method, this paper conducts an in-depth study on the application of photon input in 3D reconstruction. The study demonstrates that using non-visual information (RAW images) as input can significantly improve reconstruction accuracy, producing more accurate results compared to images processed by ISP. The paper also quantifies the impact of ISP processing on 3D reconstruction results by comparing the depth information and point cloud data of two sets of images. Experimental results show that disabling certain ISP algorithms, such as bilateral noise filtering (BNF), edge enhancement (EEH), and non-local means denoising (NLM), can further improve reconstruction accuracy and reduce errors. In conclusion, this paper proposes a photon image-based 3D reconstruction method that, combined with artificial intelligence technology and differentiable point cloud rendering techniques, holds promise for achieving higher precision and faster 3D reconstruction. This technology is of great significance in practical applications, particularly in the field of industrial close-range 3D reconstruction.

Machine Vision-Based Method for Reconstructing the Vehicle Coordinate System in End-of-Line ADAS Calibration

To ensure the accuracy and reliability of Advanced Driver Assistance Systems (ADAS), it is essential to perform offline calibration before the vehicles leave the factory. This paper proposes a method for reconstructing the vehicle coordinate system based on machine vision, which can be applied to the offline calibration of ADAS. Firstly, this study explains the preliminary preparations, such as the selection of feature points and the choice of camera model, combining actual application scenarios and testing requirements. Subsequently, the YOLO model is trained to identify and obtain feature regions, and feature point coordinates are extracted from these regions using template matching and ellipse fitting. Finally, a validation experiment is designed to evaluate the accuracy of this method using metrics such as the vehicle’s lateral and longitudinal offset distances and yaw angle. Experimental results show that, compared to traditional vehicle alignment platforms, this method improves reconstruction accuracy while reducing costs.

Low-rank tensor completion based on tensor train rank with partially overlapped sub-blocks and total variation

Recently, the low-rank tensor completion method based on tensor train (TT) rank has achieved promising performance. Ket augmentation (KA) is commonly used in TT rank-based methods to improve the performance by converting low-dimensional tensors to higher-dimensional tensors. However, block artifacts are caused since KA also destroys the original structure and image continuity of original low-dimensional tensors. To tackle this issue, a low-rank tensor completion method based on TT rank with tensor augmentation by partially overlapped sub-blocks (TAPOS) and total variation (TV) is proposed in this paper. The proposed TAPOS preserves the image continuity of the original tensor and enhances the low-rankness of the generated higher-dimensional tensors, and a weighted de-augmentation method is used to assign different weights to the elements of sub-blocks and further reduce the block artifacts. To further alleviate the block artifacts and improve reconstruction accuracy, TV is introduced in the TAPOS-based model to add the piecewise smooth prior. The parallel matrix decomposition method is introduced to estimate the TT rank to reduce the computational cost. Numerical experiments show that the proposed method outperforms the existing state-of-the-art methods.

3D-ISRNet:Generating 3D point clouds through image similarity retrieval in a complex background from a single image

Reconstructing the 3D shape of objects using limited information from a single image is a complex and necessary task. The existing methods for reconstructing point clouds from a single image haven't effectively addressed the issue of background interference in the input image, thus greatly limited their practical application. To address the underlying issue, this paper proposes a 3D-ISRNet aiming to enhance the accuracy of point cloud reconstruction through image similarity retrieval. The 3D-ISRNet retrieves point clouds similar to the object and employs spatial and channel attention mechanisms for image encoding. This aims to enhance the features of the reconstruction area and improve reconstruction accuracy. After the fusion of retrieved point clouds with image features, the combined data undergoes a symmetric encoder to generate the reconstructed point clouds, thereby alleviating the constraints imposed by the initial point clouds on the reconstruction results. 3D-ISRNet demonstrates favorable results on public and real-world datasets.

Coverage Axis++: Efficient Inner Point Selection for 3D Shape Skeletonization

AbstractWe introduce Coverage Axis++, a novel and efficient approach to 3D shape skeletonization. The current state‐of‐the‐art approaches for this task often rely on the watertightness of the input [LWS*15; PWG*19; PWG*19] or suffer from substantial computational costs [DLX*22; CD23], thereby limiting their practicality. To address this challenge, Coverage Axis++ proposes a heuristic algorithm to select skeletal points, offering a high‐accuracy approximation of the Medial Axis Transform (MAT) while significantly mitigating computational intensity for various shape representations. We introduce a simple yet effective strategy that considers shape coverage, uniformity, and centrality to derive skeletal points. The selection procedure enforces consistency with the shape structure while favoring the dominant medial balls, which thus introduces a compact underlying shape representation in terms of MAT. As a result, Coverage Axis++ allows for skeletonization for various shape representations (e.g., water‐tight meshes, triangle soups, point clouds), specification of the number of skeletal points, few hyperparameters, and highly efficient computation with improved reconstruction accuracy. Extensive experiments across a wide range of 3D shapes validate the efficiency and effectiveness of Coverage Axis++. Our codes are available at https://github.com/Frank-ZY-Dou/Coverage_Axis.

Unsupervised Feature Selection to Identify Important ICD-10 and ATC Codes for Machine Learning on a Cohort of Patients With Coronary Heart Disease: Retrospective Study.

The application of machine learning in health care often necessitates the use of hierarchical codes such as the International Classification of Diseases (ICD) and Anatomical Therapeutic Chemical (ATC) systems. These codes classify diseases and medications, respectively, thereby forming extensive data dimensions. Unsupervised feature selection tackles the "curse of dimensionality" and helps to improve the accuracy and performance of supervised learning models by reducing the number of irrelevant or redundant features and avoiding overfitting. Techniques for unsupervised feature selection, such as filter, wrapper, and embedded methods, are implemented to select the most important features with the most intrinsic information. However, they face challenges due to the sheer volume of ICD and ATC codes and the hierarchical structures of these systems. The objective of this study was to compare several unsupervised feature selection methods for ICD and ATC code databases of patients with coronary artery disease in different aspects of performance and complexity and select the best set of features representing these patients. We compared several unsupervised feature selection methods for 2 ICD and 1 ATC code databases of 51,506 patients with coronary artery disease in Alberta, Canada. Specifically, we used the Laplacian score, unsupervised feature selection for multicluster data, autoencoder-inspired unsupervised feature selection, principal feature analysis, and concrete autoencoders with and without ICD or ATC tree weight adjustment to select the 100 best features from over 9000 ICD and 2000 ATC codes. We assessed the selected features based on their ability to reconstruct the initial feature space and predict 90-day mortality following discharge. We also compared the complexity of the selected features by mean code level in the ICD or ATC tree and the interpretability of the features in the mortality prediction task using Shapley analysis. In feature space reconstruction and mortality prediction, the concrete autoencoder-based methods outperformed other techniques. Particularly, a weight-adjusted concrete autoencoder variant demonstrated improved reconstruction accuracy and significant predictive performance enhancement, confirmed by DeLong and McNemar tests (P<.05). Concrete autoencoders preferred more general codes, and they consistently reconstructed all features accurately. Additionally, features selected by weight-adjusted concrete autoencoders yielded higher Shapley values in mortality prediction than most alternatives. This study scrutinized 5 feature selection methods in ICD and ATC code data sets in an unsupervised context. Our findings underscore the superiority of the concrete autoencoder method in selecting salient features that represent the entire data set, offering a potential asset for subsequent machine learning research. We also present a novel weight adjustment approach for the concrete autoencoders specifically tailored for ICD and ATC code data sets to enhance the generalizability and interpretability of the selected features.

Research on Three-Dimensional Reconstruction of Ribs Based on Point Cloud Adaptive Smoothing Denoising.

The traditional methods for 3D reconstruction mainly involve using image processing techniques or deep learning segmentation models for rib extraction. After post-processing, voxel-based rib reconstruction is achieved. However, these methods suffer from limited reconstruction accuracy and low computational efficiency. To overcome these limitations, this paper proposes a 3D rib reconstruction method based on point cloud adaptive smoothing and denoising. We converted voxel data from CT images to multi-attribute point cloud data. Then, we applied point cloud adaptive smoothing and denoising methods to eliminate noise and non-rib points in the point cloud. Additionally, efficient 3D reconstruction and post-processing techniques were employed to achieve high-accuracy and comprehensive 3D rib reconstruction results. Experimental calculations demonstrated that compared to voxel-based 3D rib reconstruction methods, the 3D rib models generated by the proposed method achieved a 40% improvement in reconstruction accuracy and were twice as efficient as the former.

Surface temperature field real-time reconstruction of hot forging die based on 1DCNN

Accurate reconstruction of the die surface temperature field during the hot forging process is crucial for ensuring optimal control of die temperature and high-quality metal forgings. In this study, we propose a novel method that combines hot forging process parameters with a deep neural network to reconstruct the temperature field on the die surface in real-time. Firstly, a high-fidelity finite element model of the hot forging process is created and validated with a forging experiment. Secondly, temperature observation points are initially selected based on the simulated temperature distribution of the forging die, and the correlation coefficient method is used to optimize the layout of observation points to improve reconstruction accuracy and efficiency. Subsequently, multiple high-quality simulations with varying process parameters are conducted, generating a comprehensive dataset. The 1DCNN algorithm is then presented to produce a mapping from the input to the surface temperature field. Finally, the effectiveness of the surface temperature field reconstruction method is verified. The results indicate that the overall performance of the proposed 1DCNN algorithm is better than the traditional regression algorithms. In terms of reconstruction accuracy, it achieves a prediction accuracy within an absolute error of 3 °C for the outer surface points and within a relative error of 5 % for the cavity region. Regarding reconstruction efficiency, it requires an approximate execution time of 0.064 s for predicting a new sample. This paper presents a method for real-time reconstruction of the die surface temperature field, leveraging hot forging process parameters and temperature data from observation points. This approach enables reliable online monitoring of transient temperature variations on the hot forging die surface, offering valuable support for operational control and maintenance.

Elevation Resolution Enhancement Oriented 3D Ultrasound Imaging.

Three-dimensional (3D) ultrasound imaging can be accomplished by reconstructing a sequence of two-dimensional (2D) ultrasound images. However, 2D ultrasound images usually suffer from low resolution in the elevation direction, thereby impacting the accuracy of 3D reconstructed results. The lateral resolution of 2D ultrasound is known to significantly exceed the elevation resolution. By combining scanning sequences acquired from orthogonal directions, the effects of poor elevation resolution can be mitigated through a composite reconstructing process. Moreover, capturing ultrasound images from multiple perspectives necessitates a precise probe positioning method with a wide angle of coverage. Optical tracking is popularly used for probe positioning for its high accuracy and environment-robustness. In this paper, a novel large-angle accurate optical positioning method is used for enhancing resolution in 3D ultrasound imaging through orthogonal-view scanning and composite reconstruction. Experiments on two phantoms proved that our method could significantly improve reconstruction accuracy in the elevation direction of the probe compared with single-angle parallel scanning. The results indicate that our method holds the potential to improve current 3D ultrasound imaging techniques.

Learning View Synthesis for Desktop Telepresence with Few RGBD Cameras.

Recent telepresence systems have shown significant improvements in quality compared to prior systems. However, they struggle to achieve both low cost and high quality at the same time. In this work, we envision a future where telepresence systems become a commodity and can be installed on typical desktops. To this end, we present a high-quality view synthesis method that uses a cost-effective capture system that consists of commodity hardware accessible to the general public. We propose a neural renderer that uses a few RGBD cameras as input to synthesize novel views of a user and their surroundings. At the core of the renderer is Multi-Layer Point Cloud (MPC), a novel 3D representation that improves reconstruction accuracy by removing non-linear biases in depth cameras. Our temporally-aware renderer further improves the stability of synthesized videos by conditioning on past information. Additionally, we propose Spatial Skip Connections (SSC) to improve image upsampling under limited GPU memory. Experimental results show that our renderer outperforms recent methods in terms of view synthesis quality. Our method generalizes to new users and challenging content (e.g., hand gestures and clothing deformation) without costly per-video optimization, object templates, or heavy pre-processing. The code and dataset will be made available.

Reversible-Prior-Based Spectral-Spatial Transformer for Efficient Hyperspectral Image Reconstruction

The task of reconstructing a 3D cube from a 2D measurement is not well-defined in spectral imaging. Unfortunately, existing Deep Unfolding Network (DU) and End-to-End (E2E) approaches can't strike an optimal balance between computational complexity and reconstruction quality. The goal of this study is to think about ways to merge the E2E's violent mapping with DU's iterative method. Our proposed deep learning framework, the Reversible-prior-based Spectral-Spatial Transformer, combines the high-quality reconstruction capabilities of DU with the advantages of having fewer parameters and lower computing cost, similar to the E2E approach. SST-ReversibleNet uses a reversible prior to project the end-to-end mapping reconstruction results back into the measurement space, construct the residuals between the reprojection and the actual measurement, and improve reconstruction accuracy. Extensive trials show that our SST-ReversibleNet outperforms cutting-edge approaches by at least 0.8 dB and only use 34.3% Params and 44.1% giga floating-point operations per second (GFLOP).

Group sparse-based Taylor expansion method for liver pharmacokinetic parameters imaging of dynamic fluorescence molecular tomography

Objective. Pharmacokinetic parametric images obtained through dynamic fluorescence molecular tomography (DFMT) has ability of capturing dynamic changes in fluorescence concentration, thereby providing three-dimensional metabolic information for applications in biological research and drug development. However, data processing of DFMT is time-consuming, involves a vast amount of data, and the problem itself is ill-posed, which significantly limits the application of pharmacokinetic parametric images reconstruction. In this study, group sparse-based Taylor expansion method is proposed to address these problems. Approach. Firstly, Taylor expansion framework is introduced to reduce time and computational cost. Secondly, group sparsity based on structural prior is introduced to improve reconstruction accuracy. Thirdly, alternating iterative solution based on accelerated gradient descent algorithm is introduced to solve the problem. Main results. Numerical simulation and in vivo experimental results demonstrate that, in comparison to existing methods, the proposed approach significantly enhances reconstruction speed without a degradation of quality, particularly when confronted with background fluorescence interference from other organs. Significance. Our research greatly reduces time and computational cost, providing strong support for real-time monitoring of liver metabolism.

ZH3: Quadratic Zonal Harmonics

Spherical Harmonics (SH) have been used widely to represent lighting in games and film. While the quadratic (SH3) and higher order spherical harmonics represent irradiance well, they are expensive to store and evaluate, requiring 27 coefficients per sample. Linear SH (SH2), requiring only 12 coefficients, are sometimes used, but they do not represent irradiance signals accurately and can have challenges with negative reconstruction. We introduce a new representation (ZH3) that augments linear SH with just the zonal coefficient of quadratic SH, yielding significant visual improvement with just 15 coefficients, and discuss how solving for a luminance zonal axis can significantly improve reconstruction accuracy and reduce color artifacts. We also discuss how, rather than storing the ZH3 coefficients explicitly, we can hallucinate them from the linear SH, improving reconstruction accuracy over linear SH at minimal extra cost.

Bayesian model error method for the passive inverse scattering problem

This paper focuses on the passive inverse scattering problem, which uses passive measurements corresponding to randomly distributed incident sources to recover the shape of the sound-soft obstacle from a Bayesian perspective. Due to the unpredictability and randomness of incident sources, the classical Bayesian inversion framework may be unable to capture the likelihood involving the passive forward model for this inverse problem. We present the Bayesian model error method (BMEM), a novel passive imaging technique, to overcome this difficulty. The cross-correlations and the Helmholtz–Kirchhoff identity are specifically used to build an approximate active scattering model. This approximate model and the model error that it produces can be combined effectively by the suggested BMEM. The well-posedness of the posterior measure in the BMEM is proved. To further estimate the model error, an online scheme is utilized in conjunction with a preconditioned Crank–Nicolson Markov Chain Monte Carlo method to numerically approximate the posterior. Numerical experiments illustrate the effectiveness of the proposed method and also show that the online evaluation of model error can significantly improve reconstruction accuracy.

Spectral reconstruction of fundus images using retinex-based semantic spectral separation transformer, applied for retinal oximetry

Multispectral imaging, a non-invasive technique to measure retinal oxygen saturation levels, is limited due to its time-consuming and expensive nature. In this paper, we present R3ST (Retinex-based Semantic Spectral Separation Transformer), an innovative end-to-end method for RGB fundus image spectral reconstruction. Our proposed model leverages RGB images to reconstruct multispectral images, making retinal oximetry more accessible for diagnostics.Existing spectral reconstruction methods face challenges in achieving high accuracy and generalizing across camera sources, affecting diagnostic results. R3ST addresses these issues through an unsupervised semantic spectral separation module that strives to encourage the separate reconstruction of pixels with different semantics as much as possible. Additionally, we introduce an oxygen saturation loss to improve reconstruction accuracy within the relevant regions. To overcome the limitations of generalization properties, our model incorporates a self-supervised Retinex reflectance extraction module. This simulates various imaging systems and extracts reflectance information from fundus images for reconstruction.Extensive experiments demonstrate R3ST’s superior performance in spectral reconstruction accuracy, making retinal oximetry a more viable, cost-effective diagnostic option that provides essential eye health and systemic health information. The code of R3ST is available at https://github.com/telesc0pe/R3ST.