Filters








3,016 Hits in 6.0 sec

Pair-Wise Range Image Registration: A Study in Outlier Classification

Gerald Dalley, Patrick Flynn
2002 Computer Vision and Image Understanding  
In this paper, we present a robustness study on several popular techniques for performing fine registration of partially overlapping 2.5D range image pairs, with a focus on model building.  ...  The aim of these variants in both sets of tests is to classify as outliers particular point pairs containing vertices outside of the region of overlap of the two range images.  ...  In this paper, we will study the robustness of algorithms for refining the initial coarse registration when partially overlapping range image pairs are being registered.  ... 
doi:10.1006/cviu.2002.0986 fatcat:y4cqsqrerncydosrt5v2xvk24q

Range image registration for industrial inspection

Carles Matabosch, Joaquim Salvi, David Fofi, Fabrice Meriaudeau, Jeffery R. Price, Fabrice Meriaudeau
2005 Machine Vision Applications in Industrial Inspection XIII  
Building of three-dimensional models is an important topic in computer vision. Range finders only let to reconstruct a partial view of the object.  ...  In this paper, a survey of the most common techniques is presented. Furthermore experimental results are performed, and a 3D model is obtained.  ...  This paper presents a survey on range image registration. First, a classification on 3D image registration is presented in section 2.  ... 
doi:10.1117/12.585244 fatcat:av6gsay4anc2vie3sttscpkore

Better Correspondence by Registration [chapter]

Shufei Fan, Rupert Brooks, Frank P. Ferrie
2010 Lecture Notes in Computer Science  
Accurate image correspondence is crucial for estimating multiple-view geometry. In this paper, we present a registration-based method for improving accuracy of the image correspondences.  ...  Experiments show that we can improve feature localization accuracy of the MSER feature detector, which recovers the most accurate feature localization as reported in a recent study by Haja and others.  ...  For each correspondence, our registration-based refinement tries to find the optimal affine transform φ opt based on pair-wise appearances and to re-align the corresponding features accordingly.  ... 
doi:10.1007/978-3-642-12297-2_42 fatcat:6cauz6lxxvcfxkj6nwobc6quby

Hubless keypoint-based 3D deformable groupwise registration [article]

Rémi Agier, Sébastien Valette, Razmig Kéchichian, Laurent Fanton, Rémy Prost
2019 arXiv   pre-print
Our approach extracts 3D SURF keypoints from images, computes matched pairs of keypoints and registers the group by minimizing pair distances in a hubless way i.e. without computing any central mean image  ...  We present a novel algorithm for Fast Registration Of image Groups (FROG), applied to large 3D image groups.  ...  , machine learning classification and longitudinal studies.  ... 
arXiv:1809.03951v2 fatcat:h456kdvbpjb4zbchzguw43oehq

Pair-wise Registration of 3D/Color Data Sets with ICP

Lounis Douadi, Marie-jose Aldon, Andre Crosnier
2006 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems  
The purpose of the work presented in this paper is to improve performance of classical ICP. We address, here, the problem of pair-wise registration of color range images.  ...  Many variants of ICP have been proposed for the registration of 3D data sets. However, there are only a few solutions dealing with color range images.  ...  We address here the problem of pair-wise registration of dense color range images provided by a structured light sensor. This paper is organized as follows.  ... 
doi:10.1109/iros.2006.282551 dblp:conf/iros/DouadiAC06 fatcat:du7l34enabfrvkkkkbbfupxcaa

Accurate range image registration: Eliminating or modelling outliers

Yonghuai Liu, Honghai Liu, Longzhuang Li, Baogang Wei
2007 2007 IEEE Conference on Emerging Technologies & Factory Automation (EFTA 2007)  
Automatic and accurate range image registration is often a prerequisite step for range image analysis and interpretation.  ...  In this case, various techniques to eliminate and model outliers have been proposed for accurate range image registration.  ...  This shows that the explicit outlier modelling often outperforms the outlier elimination in the sense of achieving accurate automatic range image registration results, since the classification of possible  ... 
doi:10.1109/efta.2007.4416933 dblp:conf/etfa/LiuLLW07 fatcat:6qfwheqjiffqzktbesr2tl5h4q

PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency [article]

Xuyang Bai, Zixin Luo, Lei Zhou, Hongkai Chen, Lei Li, Zeyu Hu, Hongbo Fu, Chiew-Lan Tai
2021 arXiv   pre-print
Removing outlier correspondences is one of the critical steps for successful feature-based point cloud registration.  ...  In this paper, we present PointDSC, a novel deep neural network that explicitly incorporates spatial consistency for pruning outlier correspondences.  ...  Learning-based outlier rejection methods are first introduced in the 2D image matching task [48, 78, 79, 65] , where outlier rejection is formulated as an inlier/outlier classification problem.  ... 
arXiv:2103.05465v1 fatcat:ksw465ecvnc6zd6yxf52iuzkca

CORSAIR: Convolutional Object Retrieval and Symmetry-AIded Registration [article]

Tianyu Zhao, Qiaojun Feng, Sai Jadhav, Nikolay Atanasov
2021 arXiv   pre-print
Our model extends the Fully Convolutional Geometric Features model to learn a global object-shape embedding in addition to local point-wise features from the point-cloud observations.  ...  Our formulation also leverages symmetries, present in the object shapes, to obtain promising local-feature pairs from different symmetry classes for matching.  ...  INTRODUCTION Advances in learning-based computer vision algorithms have enabled detection, classification, and segmentation of objects in images and videos with impressive accuracy.  ... 
arXiv:2103.06911v3 fatcat:gjnoiylz6belxmt6374rujg7wm

Adding Chaos to Differential Evolution for Range Image Registration [chapter]

Ivanoe De Falco, Antonio Della Cioppa, Domenico Maisto, Umberto Scafuri, Ernesto Tarantino
2013 Lecture Notes in Computer Science  
This paper presents a method for automatically pair-wise registering range images.  ...  Registration is effected adding chaos to a Differential Evolution technique and by applying the Grid Closest Point algorithm to find the best possible transformation of the second image causing 3D reconstruction  ...  A complete study focused on pair-wise registration is presented in [5] , while different techniques for both pair-wise and multi-view registration, and a new classification, can be found in [21] .  ... 
doi:10.1007/978-3-642-37192-9_35 fatcat:2exxkg4y4raxhjyhe7ku5euski

A review of recent range image registration methods with accuracy evaluation

Joaquim Salvi, Carles Matabosch, David Fofi, Josep Forest
2007 Image and Vision Computing  
Furthermore, a sample of the techniques has been programmed and experimental results are reported to determine the best method in the presence of noise and outliers, providing a useful guide for an interested  ...  The first problem is related to obtaining a rough registration when such motion is not available. The second one is focused on obtaining a fine registration from an initial approximation.  ...  The authors thank also Timo Zinsser and Szymon Rusinkiewicz for providing the synthetic images used in their papers.  ... 
doi:10.1016/j.imavis.2006.05.012 fatcat:rvqp6v3rsjhmdnrifiy3vcevmu

SVF-Net: Learning Deformable Image Registration Using Shape Matching [chapter]

Marc-Michel Rohé, Manasi Datar, Tobias Heimann, Maxime Sermesant, Xavier Pennec
2017 Lecture Notes in Computer Science  
Not only our method is more accurate but it is also faster -registration of two 3D-images taking less than 30ms second on a GPU -and more robust to outliers.  ...  In this paper, we propose an innovative approach for registration based on the deterministic prediction of the parameters from both images instead of the optimization of a energy criteria.  ...  In our application, we study cardiac MRI short-axis images, which are acquired with a non-isotropic resolution in the Z axis (slice spacing ranges from 6 to 10mm whereas in-plane spacing ranges from 1.5  ... 
doi:10.1007/978-3-319-66182-7_31 fatcat:44jz7hsiu5bl3buzib2pnen4da

Shape comparison using perturbing shape registration

Yifeng Jiang, Erin Edmiston, Fei Wang, Hilary P. Blumberg, Lawrence H. Staib, Xenophon Papademetris
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
Shape registration is often involved in computing statistical differences between groups of shapes, which is a key aspect of morphometric study.  ...  This paper proposes a perturbation scheme, which perturbs registrations by feeding them with different resampled shape groups, and then aggregates the resulting shape differences.  ...  This work was supported by the NIH under grants R01EB006494 (XP), R01MH070902 (HPB), R01MH069747 (HPB), R01NS35193 (LHS, XP), and the National Alliance for Research in Schizophrenia and Depression (FW,  ... 
doi:10.1109/cvpr.2009.5206598 dblp:conf/cvpr/JiangEWBSP09 fatcat:2kt5zu7epva3dmaunxmhshfgxy

Shape comparison using perturbing shape registration

Yifeng Jiang, E. Edmiston, Fei Wang, H.P. Blumberg, L.H. Staib, X. Papademetris
2009 2009 IEEE Conference on Computer Vision and Pattern Recognition  
Shape registration is often involved in computing statistical differences between groups of shapes, which is a key aspect of morphometric study.  ...  This paper proposes a perturbation scheme, which perturbs registrations by feeding them with different resampled shape groups, and then aggregates the resulting shape differences.  ...  This work was supported by the NIH under grants R01EB006494 (XP), R01MH070902 (HPB), R01MH069747 (HPB), R01NS35193 (LHS, XP), and the National Alliance for Research in Schizophrenia and Depression (FW,  ... 
doi:10.1109/cvprw.2009.5206598 pmid:20386618 pmcid:PMC2852275 fatcat:cvh2md5v4jcy7cyjcn2sc4vtqa

Group-wise similarity registration of point sets using Student's t-mixture model for statistical shape models

Nishant Ravikumar, Ali Gooya, Serkan Çimen, Alejandro F. Frangi, Zeike A. Taylor
2018 Medical Image Analysis  
A probabilistic group-wise similarity registration technique based on Student's t-mixture model (TMM) and a multi-resolution extension of the same (mr-TMM) are proposed in this study, to robustly align  ...  Automated segmentation and landmarking of medical images often result in shapes with varying proportions of outliers and consequently require a robust method of alignment and correspondence estimation.  ...  Acknowledgements This study was funded by the European Unions Seventh Framework Programme ( FP7 /2007 2013) as part of the project VPH-DARE@IT (grant agreement no. 601055 ), and by the Engineering and  ... 
doi:10.1016/j.media.2017.11.012 pmid:29248842 fatcat:jcallum4vfdcnnbdgy2jqrwslu

Deep Hough Voting for Robust Global Registration [article]

Junha Lee, Seungwook Kim, Minsu Cho, Jaesik Park
2021 arXiv   pre-print
Point cloud registration is the task of estimating the rigid transformation that aligns a pair of point cloud fragments.  ...  First, deep geometric features are extracted from a point cloud pair to compute putative correspondences.  ...  A.4. Additional scene-wise statistics In Table a .3, the comprehensive scene-wise statistics for our best-performing model on the 3DMatch benchmark are presented.  ... 
arXiv:2109.04310v1 fatcat:s45vb7if5vgsno3fbcmxeyteha
« Previous Showing results 1 — 15 out of 3,016 results