检索结果-内蒙古大学图书馆

conference on Computer Vision and Pattern Recognition (CVPR)

作者： Julian Chibane Thiemo Alldieck Gerard Pons-Moll Max Planck Institute for Informatics Saarland Informatics Campus Germany University of Würzburg Germany Computer Graphics Lab TU Braunschweig Germany

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

While many works focus on 3D reconstruction from images, in this paper, we focus on 3D shape reconstruction and completion from a variety of 3D inputs, which are deficient in some respect: low and high resolution voxels, sparse and dense point clouds, complete or incomplete. Processing of such 3D inputs is an increasingly important problem as they are the output of 3D scanners, which are becoming more accessible, and are the intermediate output of 3D computer vision algorithms. Recently, learned implicit functions have shown great promise as they produce continuous reconstructions. However, we identified two limitations in reconstruction from 3D inputs: 1) details present in the input data are not retained, and 2) poor reconstruction of articulated humans. To solve this, we propose Implicit Feature Networks (IF-Nets), which deliver continuous outputs, can handle multiple topologies, and complete shapes for missing or sparse input data retaining the nice properties of recent learned implicit functions, but critically they can also retain detail when it is present in the input data, and can reconstruct articulated humans. Our work differs from prior work in two crucial aspects. First, instead of using a single vector to encode a 3D shape, we extract a learnable 3-dimensional multi-scale tensor of deep features, which is aligned with the original Euclidean space embedding the shape. Second, instead of classifying x-y-z point coordinates directly, we classify deep features extracted from the tensor at a continuous query point. We show that this forces our model to make decisions based on global and local shape structure, as opposed to point coordinates, which are arbitrary under Euclidean transformations. Experiments demonstrate that IF-Nets outperform prior work in 3D object reconstruction in ShapeNet, and obtain significantly more accurate 3D human reconstructions. Code and project website is available at https://***/ifnets/.

关键词： Three-dimensional displays Shape image reconstruction Topology Feature extraction Computer vision Tensile stress

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral image Recovery Using a Color Camera for Detecting Colonies of Foodborne Pathogens on Agar Plate

引用

Journal of Biosystems Engineering 2019年第3期44卷 169-185页

作者： Yoon, Seung-Chul Shin, Tae-Sung Heitschmidt, Gerald W. Lawrence, Kurt C. Park, Bosoon Gamble, Gary U.S. Department of Agriculture Agricultural Research Service U.S. National Poultry Research Center 950 College Station Rd. AthensGA30605 United States

Purpose: Hyperspectral imaging often requires a special camera system to obtain spectral images. The cost for acquisition and process of hyperspectral images is usually much higher than color images. On the contrary, typical consumer-grade digital color cameras are much cheaper to obtain and process spatially high-resolution images than hyperspectral cameras. This paper is concerned with the development of a hyperspectral image recovery technique that can reconstruct hyperspectral images only from color images obtained by a digital color camera. Methods: A sparse representation and least squares regression-based classification of foodborne pathogens on agar plates are presented. The target pathogen bacteria were the six representative non-O157 Shiga toxin-producing Escherichia coli (STEC) serogroups (O26, O45, O103, O111, O121, and O145). The wavelength range for the spectral recovery was from 400 to 1000 nm. Unlike many other studies using color charts with known and noise-free spectra for training their spectral recovery models, we directly used the hyperspectral and color images of real scenes for training the spectral recovery models. Both hyperspectral and color images were calibrated to percent reflectance values and then spatially registered. Two spectral recovery models including polynomial multivariate linear regression (MLR) and partial least squares regression (PLSR) were evaluated and compared by cross-validation and independent test. Results: The spectral recovery results showed that the PLSR was more effective than the MLR. The classification accuracy measured with recovered spectra in an independent test set was about 5–10% less than the case of using the true hyperspectral images although the difference in maximum classification accuracy was only about 2%. Conclusion: The results suggested the potential of a cost-effective color imaging system using hyperspectral image classification algorithms for differentiating pathogens in agar plates. © 2019, Th

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Model Assessment Through data Assimilation of Realistic data in Cardiac Electrophysiology 10th

Model Assessment Through Data Assimilation of Realistic Data...

引用

10th International conference on Functional Imaging and Modeling of the Heart (FIMH)

作者： Gerard, Antoine Collin, Annabelle Bureau, Gautier Moireau, Philippe Coudiere, Yves Electrophysiol & Heart Modeling Inst IHU Liryc F-33000 Pessac France Univ Bordeaux IMB UMR 5251 Talence France Bordeaux INP IMB UMR 5251 Talence France INRIA Inria Bordeaux Sud Ouest Talence France Univ Paris Saclay INRIA Palaiseau France Ecole Polytech CNRS UMR 7649 Inst Polytech Paris LMS Palaiseau France

ISBN: (纸本)9783030219499;9783030219482

We consider a model-based estimation procedure - namely a data assimilation algorithm - of the atrial depolarization state of a subject using data corresponding to electro-anatomical maps. Our objective is to evaluate the sensitivity of such a model-based reconstruction with respect to model choices. The followed data assimilation approach is capable of using electrical activation times to adapt a monodomain model simulation, thanks to an ingenious model-data fitting term inspired from image processing. The resulting simulation smoothes and completes the activation maps when they are spatially incomplete. Moreover, conductivity parameters can also be inferred. The model sensitivity assessment is performed based on synthetic data generated with a validated realistic atria model and then inverted using simpler modeling ingredients. In particular, the impact of the muscle fibers definition and corresponding anisotropic conductivity parameters is studied. Finally, an application of the method to real data is presented, showing promising results.

关键词： Cardiac modeling Electrophysiology data assimilation

来源：评论

学校读者我要写书评

暂无评论

Recovery of Subspace Structure from High-Rank data with Missing Entries

Recovery of Subspace Structure from High-Rank Data with Miss...

引用

IEEE International conference on image Processing

作者： João Carvalho Manuel Marques João P. Costeira Univ. Lisboa Lisbon Portugal

We propose a method to reconstruct and cluster incomplete high-dimensional data lying in a union of low-dimensional subspaces. Exploring the sparse representation model, we jointly estimate the missing data while imposing the intrinsic subspace structure. Although we have a non-convex problem, we propose an algorithm robust to initialization. Extensive experiments with synthetic and real data show that our approach leads to significant improvements in the reconstruction and segmentation, outperforming current state of the art for both low and high-rank data.

关键词： Clustering algorithms Sparse matrices image reconstruction data models Motion segmentation Computer vision Signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

Sinogram interpolation method for limited-angle tomography

Sinogram interpolation method for limited-angle tomography

引用

International conference on Advanced Electrical Engineering (ICAEE)

作者： A. Benammar A. Allag R. Drai M. yahi T. Boutkedjirt Research Center in Industrial Technologies (CRTI) Division of signal processing and imagery Cheraga Algiers Algeria Faculty of Physics El-Alia University of Sciences and Technology Houari Boumediene Algiers Algeria

ISBN: (数字)9781728122205

ISBN: (纸本)9781728122212

This work aims to study and implement a reconstruction method of X-ray tomography able to reconstruct the image from insufficent number of projections. We proposed a new algorithm using the sinogram interpolation. The sinogram restoration is based on the incomplete sinogram available and generates additional data. The reconstructed images were obtained from the interpolated projections, using the classical FBP method, the quality of the reconstructed image depend on the quality and quantity of the projections. Our study showed image restoration performance in case of missing data and shows improved quality of CT images with reduced artifacts in the reconstruction results.

关键词： image reconstruction Interpolation image restoration Computed tomography reconstruction algorithms image quality

来源：评论

学校读者我要写书评

暂无评论

Motion image data collection simulation for structure-from-motion 3D target reconstruction 8

Motion image data collection simulation for structure-from-m...

引用

conference on Geospatial Informatics, Motion imagery, and Network Analytics viii

作者： Kalukin, Andrew Pinkerton, Robert Natl Geospatial Intelligence Agcy 7500 GEOINT Dr Springfield VA 22150 USA

ISBN: (数字)9781510618022

ISBN: (纸本)9781510618022

Motion imagery can be used in some circumstances as a source of data for reconstructing three-dimensional (3D) representations of targets and objects in the scene. This research explores the utility of using motion image data collection simulation for input to a 3D target reconstruction algorithm based on structure-from-motion. The use of simulated video is advantageous for testing potential collections on targets or geographical areas for which real video and images may be unavailable. Examples are provided of tests of angular sampling and occlusion, degradation of input imagery through blurring, and measurement of their effects on 3D reconstruction quality.

关键词： Structure-from-Motion 3D reconstruction video collection simulation video SfM motion imagery

来源：评论

学校读者我要写书评

暂无评论

Reconstruct missing pixels of Landsat land surface temperature product using a CNN with partial convolution

Reconstruct missing pixels of Landsat land surface temperatu...

引用

conference on Applications of Machine Learning

作者： Chen, Maosi Newell, Benjamin H. Sun, Zhibin Corr, Chelsea A. Gao, Wei USDA UVB Monitoring & Res Program Nat Resource Ecol Lab Ft Collins CO 80521 USA Colorado State Univ Geospatial Centroid Ft Collins CO 80523 USA Colorado State Univ Dept Ecosyst Sci & Sustainabil Ft Collins CO 80523 USA

ISBN: (纸本)9781510629721

U.S. Landsat Analysis Ready data (ARD) recently included the Land Surface Temperature (LST) product, which contains widespread and irregularly-shaped missing pixels due to cloud contamination or incomplete satellite coverage. Many analyses rely on complete LST images therefore techniques that accurately fill data gaps are needed. Here, the development of a partial-convolution based model with the U-Net like architecture to reconstruct the missing pixels in the ARD LST images is discussed. The original partial convolution layer is modified to consider both the convolution kernel weights and the number of valid pixels in the calculation of the mask correction ratio. In addition, the new partial merge layer is developed to merge feature maps according to their masks. Pixel reconstruction using this model was conducted using Landsat 8 ARD LST images in Colorado between 2014 and 2018. Complete LST patches (64x64) for two identical scenes acquired at different dates (up to 48 days apart) were randomly paired with ARD cloud masks to generate the model inputs. The model was trained for 10 epochs and the validation results show that the average RMSE values for a restored LST image in the unmasked, masked, and whole region are 0.29K, 1.00K, and 0.62K, respectively. In general, the model is capable of capturing the high-level semantics from the inputs and bridging the difference in acquisition dates for gap filling. The transition between the masked and unmasked regions (including the edge area of the image) in restored images is smooth and reflects realistic features (e.g., LST gradients). For large masked areas, the reference provides semantics at both low and high levels.

关键词： partial convolution layer partial merge layer satellite image inpainting Landsat Analysis Ready data (ARD) Land Surface Temperature (LST)

来源：评论

学校读者我要写书评

暂无评论

Gait Energy image Restoration Using Generative Adversarial Networks

Gait Energy Image Restoration Using Generative Adversarial N...

引用

IEEE International conference on image Processing

作者： Maryam Babaee Yue Zhu Okan Köpüklü Stefan Hörmann Gerhard Rigoll Tech. Univ. Munich Munich Germany

Gait is a biometric property that can be used for human identification in video surveillance. Basically, different gait features require motion of a person walking over one complete gait cycle. For example, in Gait Energy image (GEI), average of silhouette images over one complete gait cycle is computed. However, in reality, there might be a partial gait cycle data available due to occlusion. In this paper, we propose a Generative Adversarial Network (GAN) in order to address the problem of gait recognition from incomplete gait cycle. Precisely, the network is able to reconstruct complete GEIs from incomplete GEIs. The proposed architecture is composed of (i) a generator which is an auto-encoder network to construct complete GEIs out of incomplete GEIs and (ii) two discriminators, one of which discriminates whether a given image is a full GEI while the other discriminates whether two GEIs belong to the same subject. We evaluate our approach on the OULP large gait dataset confirming that the proposed architecture successfully reconstructs complete GEIs from even extreme incomplete gait cycles.

关键词： image reconstruction Generators Gait recognition Legged locomotion Generative adversarial networks Training Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Efficient Depth image reconstruction using Accelerated Proximal Gradient Method 10

Efficient Depth Image Reconstruction using Accelerated Proxi...

引用

10th International conference on Knowledge and Systems Engineering (KSE)

作者： Mau Uyen Nguyen Thanh Tinh Dao Van Ha Tang Le Quy Don Tech Univ Fac Informat Technol Hanoi Vietnam

ISBN: (纸本)9781538661130

This paper presents an accelerated proximal gradient technique for depth image reconstruction from incomplete data samples. One of the most technical challenges in depth imaging is to estimate a depth image from sparse measurements obtained due to missing depth values or fast sensing. Recent advances in signal processing, i.e., compressive sensing (CS) allows the depth image to be reconstructed precisely from far reduced measurements provided that the image has sparse representations in proper bases. Inspired by the CS theory, this paper formulates the task of depth channel reconstruction as a sparsity-regularized least squares optimization problem. To solve this problem efficiently, an iterative algorithm based on the accelerated proximal gradient technique is developed, which not only speeds up the convergence rate, but also enhances the quality of depth image estimation. Several experiments are conducted and the results confirm the efficiency of the proposed approach.

关键词： image reconstruction Convergence Acceleration Optimization Robot sensing systems Task analysis

来源：评论

学校读者我要写书评

暂无评论

SEM Annual conference and Exposition on Experimental and Applied Mechanics, 2018

SEM Annual Conference and Exposition on Experimental and App...

引用

36th IMAC, A conference and Exposition on Structural Dynamics, 2018

ISBN: (纸本)9783319950730

The proceedings contain 120 papers. The special focus in this conference is on Exposition on Experimental and Applied Mechanics. The topics include: Inverse identification of the loading applied by a tire on a landing gear wheel;Fatigue properties assessment of API 5l Gr. B pipeline steel using infrared thermography;determination of constitutive parameters in inverse problem using thermoelastic data;experimental investigation of emissivity influence to obtain thermal field by near infrared thermography;experimental validation of the energy balance equation in the presence of acoustic emission;model based inversion for pulse thermography;experimentally enhanced computations: Calibration methodology for an anisotropic metal, Part I: Traditional approach;Mechanical response and energy stored during deformation of crystallizing TPU;Measuring strain-induced crystallinity in rubbers from IR thermography;chapter 12 mechanical and thermomechanical characterization of different leathers;Experimentally enhanced computations: Calibration methodology for an anisotropic metal, Part II – Novel approach/validation;multi-instrumentation of very high temperature tests;detection of damage during quasi-static loading of a single stringer panel using passive thermography and acoustic emission;The effect of chamber temperature on residual stresses of FDM parts;system identification of structures with incomplete modal information;towards integrating imaging techniques to assess manufacturing features and in-service damage in composite components;image-based stress field reconstruction in complex media;infrared thermography for material characterization at intermediate strain rates;evaluation of fatigue damage in short carbon fiber reinforced plastics based on thermoelastic stress and phase analysis;thermographic stress field investigation of a multiply-loaded disk;thermoelastic measurement techniques enabled by self-reference.

关键词： Thermography (imaging)

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：