2,184 Hits in 9.2 sec

A new vector field distance transform and its application to mesh processing from 3D scanned data

Marc Fournier, Jean-Michel Dischler, Dominique Bechmann
2007 The Visual Computer  
In this paper we define a new 3D vector field distance transform to implicitly represent a mesh surface.  ...  Widely used Marching Cube triangulation algorithm is adapted to the new vector field distance transform to correctly reconstruct the resulting explicit surface.  ...  Acknowledgment We wish to thank the following sources for the use of the models included in this paper:  ... 
doi:10.1007/s00371-007-0143-6 fatcat:gtsjusdztfcujjddxlmaoi7rwa

Generation of micro-scale finite element models from synchrotron X-ray CT images for multidirectional carbon fibre reinforced composites

R.M. Sencu, Z. Yang, Y.C. Wang, P.J. Withers, C. Rau, A. Parson, C. Soutis
2016 Composites. Part A, Applied science and manufacturing  
The new algorithm uses a global overlapping stack filtering step followed by a local fibre tracking step. Both steps are based on the Bayesian inference theory.  ...  are then used to generate micro-scale finite element models.  ...  These were implemented by an inverse exponential fitting to find the adaptive threshold coefficients using Fig. 10 .  ... 
doi:10.1016/j.compositesa.2016.09.010 fatcat:pjgaulyfpnalhagcgv3y7423nm

Investigation of Smoothness-Increasing Accuracy-Conserving Filters for Improving Streamline Integration through Discontinuous Fields

M. Steffen, S. Curtis, R.M. Kirby, J.K. Ryan
2008 IEEE Transactions on Visualization and Computer Graphics  
We investigate whether such an approach applied to uniform quadrilateral discontinuous Galerkin (high-order finite volume) data can be used to augment current adaptive error control approaches.  ...  As the root of the difficulties is the discontinuous nature of the data, we present a complementary approach of applying smoothness-increasing accuracy-conserving filters to the data prior to streamline  ...  ACKNOWLEDGMENTS The first author would like to acknowledge support from the US Department of Energy through the Center for the Simulation of Accidental Fires and Explosions (C-SAFE) under Grant W-7405-  ... 
doi:10.1109/tvcg.2008.9 pmid:18369273 fatcat:dz4b5o4p6nhijcz7soggmqleue

Cross-Atlas Convolution for Parameterization Invariant Learning on Textured Mesh Surface

Shiwei Li, Zixin Luo, Mingmin Zhen, Yao Yao, Tianwei Shen, Tian Fang, Long Quan
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
We present a convolutional network architecture for direct feature learning on mesh surfaces through their atlases of texture maps.  ...  Since the parameterization of texture map is not pre-determined, and depends on the surface topologies, we therefore introduce a novel cross-atlas convolution to recover the original mesh geodesic neighborhood  ...  Receptive fields Although the convolution is applied over 2D texture maps, its receptive field follows the mesh geodesic distance. This is done by using the offset map to redirect pixel locations.  ... 
doi:10.1109/cvpr.2019.00630 dblp:conf/cvpr/LiLZ0SFQ19 fatcat:gx7cj37d6zdcbbq3igyyonfa7a

Feature-Preserving Volume Filtering [article]

László Neumann, Balázs Csébfalvi, Ivan Viola, Matej Mlejnek, Eduard Gröller
2002 EUROVIS 2005: Eurographics / IEEE VGTC Symposium on Visualization  
Previous global 3D methods are restricted to binary volumes or segmented iso-surfaces and they are based on area minimization of one single reconstructed surface.  ...  Although the strength of the presented algorithm is demonstrated on a specific 2D and a specific 3D application, it is considered as a general mathematical tool for processing images and volumes.  ...  The 2D and 3D convolution kernels  ... 
doi:10.2312/vissym/vissym02/105-114 fatcat:glbfo7a3x5gptp3tozv77jwxzu

From Feature Detection in Truncated Signed Distance Fields to Sparse Stable Scene Graphs

Daniel R. Canelhas, Todor Stoyanov, Achim J. Lilienthal
2016 IEEE Robotics and Automation Letters  
A particularly important representation is the truncated signed distance field (TSDF) that is at the core of recent advances in dense 3D mapping.  ...  We motivate our study with an example application for building sparse stable scene graphs, and present an efficient GPU-parallel algorithm to obtain the graphs, made possible by the combination of TSDF  ...  Generating the volumetric filter kernels from the 1D coefficient vectors is analogous to the 2D case and is detailed in Algorithm 1.  ... 
doi:10.1109/lra.2016.2523555 dblp:journals/ral/CanelhasSL16 fatcat:owilo4lfmfa65lsusyeultj2ay

Lines of Curvature for Polyp Detection in Virtual Colonoscopy

Lingxiao Zhao, Charl Botha, Javier Bescos, Roel Truyen, Frans Vos, Frits Post
2006 IEEE Transactions on Visualization and Computer Graphics  
We have adapted existing approaches on explicit triangular surface meshes, and developed a new algorithm on implicit surfaces embedded in 3D volume data.  ...  Most of the previous work on automatic polyp detection makes use of indicators based on the scalar curvature of the colon wall and can result in many false-positive detections.  ...  Frans A. Gerritsen and Ir. C. van Wijk for kind advices and helpful discussions, as well as providing the authors with CT scans of patients.  ... 
doi:10.1109/tvcg.2006.158 pmid:17080813 fatcat:wpjwcsf6qjcythw3dv3w5no5lu

A Non-linear Differential CNN-Rendering Module for 3D Data Enhancement [article]

Yonatan Svirsky, Andrei Sharf
2019 arXiv   pre-print
Our module is learnable and can be easily integrated into neural networks allowing to optimize data rendering towards specific learning tasks using gradient based methods in an end-to-end fashion.  ...  In our experiments, we apply our module to demonstrate efficient localization and classification tasks in cluttered data both 2D and 3D.  ...  Similar to us, ProbNet [23] use a sparse set of probes in 3D space and a volume distance field the sense shape geometry features.  ... 
arXiv:1904.04850v1 fatcat:4neyi6oh7jhzzjvfmeboxotucm

VoxSegNet: Volumetric CNNs for Semantic Part Segmentation of 3D Shapes [article]

Zongji Wang, Feng Lu
2018 arXiv   pre-print
Voxel is an important format to represent geometric data, which has been widely used for 3D deep learning in shape analysis due to its generalization ability and regular data format.  ...  In this paper, we propose a novel volumetric convolutional neural network, which could extract discriminative features encoding detailed information from voxelized 3D data under a limited resolution.  ...  [29] presented a learning-based method to segment and label 3D meshes by combining various geometric features. Later, Guo et al.  ... 
arXiv:1809.00226v1 fatcat:rlxoahgzdvdcbbdmv3j453tp2q

A Review on Deep Learning Approaches for 3D Data Representations in Retrieval and Classifications

Abubakar Sulaiman Gezawa, Yan Zhang, Qicong Wang, Lei Yunqi
2020 IEEE Access  
Therefore, it can be concluded that deep learning together with a suitable 3D data representation gives an effective approach for improving the performance of 3D shape analysis.  ...  However, implementing the methods in 3D data is a bit complex because most of the previously designed deep learning architectures used 1D or 2D as input.  ...  In [4] , a voxel-based method was used to characterize a 3D shape and use a 3D CNN to the whole volume. References [79] , [80] used the features describe on a manifold to execute CNN operations.  ... 
doi:10.1109/access.2020.2982196 fatcat:jnya5rscynf3zm7efuucqxafri

Segmentation of distal airways using structural analysis

Debora Gil, Carles Sanchez, Agnes Borras, Marta Diez-Ferrer, Antoni Rosell, Niels Bergsland
2019 PLoS ONE  
We have compared our method to the top performers of the EXACT09 challenge and to a commercial software for biopsy planning evaluated in an own-collected data-base of high resolution CT scans acquired  ...  Results on our data-base show the reliability across varying breathing conditions and a competitive performance for biopsy planning compared to a commercial solution.  ...  Acknowledgments Dedicated to "la banquera solidaria, mi Heidi alcoholica y los suculentos biberones veraniegos". Visualization: Agnes Borras. Writing -original draft: Debora Gil.  ... 
doi:10.1371/journal.pone.0226006 pmid:31856216 pmcid:PMC6922352 fatcat:u2k22pv6hzfyngxxtni65ay3fe

Analytical solutions for sketch-based convolution surface modeling on the GPU

Xiaoqiang Zhu, Xiaogang Jin, Shengjun Liu, Hanli Zhao
2011 The Visual Computer  
For finite support kernel functions, a skeleton clipping algorithm is presented to compute the valid skeletons.  ...  We convert the double integral over a planar polygon into a simple integral along the contour of the polygon based on Green's theorem, which reduces the computational cost and allows for efficient parallel  ...  A convolution surface is defined by convolving a skeleton with a three-dimensional low-pass filter kernel. The skeletal elements can be points, polylines, polygons, and even volumes.  ... 
doi:10.1007/s00371-011-0662-z fatcat:2pflrrli5zfd5mxah2bwkxfx6m

Real-time Rendering of Stack-based Terrains [article]

Falko Löffler, Andreas Müller, Heidrun Schumann
2011 International Symposium on Vision, Modeling, and Visualization  
In contrast, a 3D data representation quickly exceeds the available memory resources. To overcome this problem we apply material stacks.  ...  Usually, terrain rendering relies on a 2D regular grid of height values, the so called height field. Height fields describe 2.5D surfaces and are not able to present complex 3D terrain features.  ...  For 3D terrain data similar data-structures can be applied. For instance, in [Gei07, GT07] an octree is used to manage 3D terrain volume.  ... 
doi:10.2312/pe/vmv/vmv11/161-168 dblp:conf/vmv/LofflerMS11 fatcat:6u6a4gvapnfjbjwbybspvbquwy

3D Model Generation and Reconstruction Using Conditional Generative Adversarial Network

Haisheng Li, Yanping Zheng, Xiaoqun Wu, Qiang Cai
2019 International Journal of Computational Intelligence Systems  
Moreover, to better guide generator to reconstruct 3D model from a single image in high quality, we propose a new 3D model reconstruction network by integrating a classifier into the traditional system  ...  To address this problem, we add the class information to both generator and discriminator and construct a new network named 3D conditional GAN.  ...  a Single Image If a 3D model of a specified condition can be generated from a random vector, it's possible to generate a 3D model corresponding to an image by encoding the image and using encoded result  ... 
doi:10.2991/ijcis.d.190617.001 fatcat:mkkkbwi7sbbwjeukvinbrwplwy

Learning Depth-Guided Convolutions for Monocular 3D Object Detection

Mingyu Ding, Yuqi Huo, Hongwei Yi, Zhe Wang, Jianping Shi, Zhiwu Lu, Ping Luo
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
To better represent 3D structure, prior arts typically transform depth maps estimated from 2D images into a pseudo-LiDAR representation, and then apply existing 3D point-cloud based object detectors.  ...  LCN (D 4 LCN), where the filters and their receptive fields can be automatically learned from image-based depth maps, making different pixels of different images have different filters.  ...  Since there are huge intra-class and inter-class scale differences in an RGB image, we use I to learn an adaptive dilation rate for each filter to obtain different sizes of receptive fields by an adaptive  ... 
doi:10.1109/cvpr42600.2020.01169 dblp:conf/cvpr/DingHYWSLL20a fatcat:uol5jq2n25gwjdzghz7eke63i4
« Previous Showing results 1 — 15 out of 2,184 results