P2MAT-NET: Learning medial axis transform from sparse point clouds

Abstract The medial axis transform (MAT) of a 3D shape includes the set of centers and radii of the maximally inscribed spheres, and is a complete shape descriptor that can be used to reconstruct the original shape. It is a compact representation that jointly describes geometry, topology, and symmetry properties of a given shape. In this work, we present P2MAT-NET, a neural network which learns the pattern of sparse point clouds and transform them into spheres approximating MAT. The experimental results illustrate that P2MAT-NET demonstrates better performance than state-of-the-art methods in computing MAT from point clouds, in terms of MAT quality to approximate the 3D shapes. The computed MAT can be used as an intermediate descriptor for downstream applications such as 3D shape recognition from point clouds. Our results show that it can achieve competitive performance in recognition with state-of-the-art methods.

[1]  Andrea Tagliasacchi,et al.  3D Skeletons: A State‐of‐the‐Art Report , 2016, Comput. Graph. Forum.

[2]  Tao Ju,et al.  Variational implicit point set surfaces , 2019, ACM Trans. Graph..

[3]  Szymon Rusinkiewicz,et al.  Learning Detail Transfer based on Geometric Features , 2017, Comput. Graph. Forum.

[4]  Jianxiong Xiao,et al.  3D ShapeNets: A deep representation for volumetric shapes , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Bin Wang,et al.  Q-MAT+: An error-controllable and feature-sensitive simplification algorithm for medial axis transform , 2019, Comput. Aided Geom. Des..

[6]  Raquel Urtasun,et al.  Deep Parametric Continuous Convolutional Neural Networks , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[7]  Yue Wang,et al.  Dynamic Graph CNN for Learning on Point Clouds , 2018, ACM Trans. Graph..

[8]  Tamy Boubekeur,et al.  Sphere-Meshes , 2013, ACM Trans. Graph..

[9]  Bin Wang,et al.  MAT-Net: Medial Axis Transform Network for 3D Object Recognition , 2019, IJCAI.

[10]  Michael M. Kazhdan,et al.  Screened poisson surface reconstruction , 2013, TOGS.

[11]  Gabriel Taubin,et al.  A benchmark for surface reconstruction , 2013, TOGS.

[12]  Andrea Tagliasacchi,et al.  LSMAT Least Squares Medial Axis Transform , 2019, Comput. Graph. Forum.

[13]  Yizhou Yu,et al.  Medial Meshes – A Compact and Accurate Representation of Medial Axis Transform , 2016, IEEE Transactions on Visualization and Computer Graphics.

[14]  Michael M. Kazhdan,et al.  Poisson surface reconstruction , 2006, SGP '06.

[15]  Jiaxin Li,et al.  SO-Net: Self-Organizing Network for Point Cloud Analysis , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[16]  Daniel Cohen-Or,et al.  EC-Net: an Edge-aware Point set Consolidation Network , 2018, ECCV.

[17]  Marshall W. Bern,et al.  Surface Reconstruction by Voronoi Filtering , 1998, SCG '98.

[18]  Daniel Cohen-Or,et al.  Patch-Based Progressive 3D Point Set Upsampling , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  Jonathan Masci,et al.  Geometric Deep Learning on Graphs and Manifolds Using Mixture Model CNNs , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Anath Fischer,et al.  Nesti-Net: Normal Estimation for Unstructured 3D Point Clouds Using Convolutional Neural Networks , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Xiaoguang Han,et al.  A Skeleton-Bridged Deep Learning Approach for Generating Meshes of Complex Topologies From Single RGB Images , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Leonidas J. Guibas,et al.  PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[23]  Leonidas J. Guibas,et al.  PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space , 2017, NIPS.

[24]  Caiming Zhang,et al.  Q-MAT , 2015, ACM Trans. Graph..

[25]  Sang Won Bae,et al.  3D medial axis point approximation using nearest neighbors and the normal field , 2011, The Visual Computer.

[26]  Yang Liu,et al.  O-CNN , 2017, ACM Trans. Graph..

[27]  Xiaohu Guo,et al.  DMAT: Deformable Medial Axis Transform for Animated Mesh Approximation , 2018, Comput. Graph. Forum.

[28]  Daniel Cohen-Or,et al.  PU-Net: Point Cloud Upsampling Network , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[29]  Kaleem Siddiqi,et al.  Local Spectral Graph Convolution for Point Set Feature Learning , 2018, ECCV.

[30]  Daniel Cohen-Or,et al.  P2P-NET , 2018, ACM Trans. Graph..

[31]  Maks Ovsjanikov,et al.  PCPNet Learning Local Shape Properties from Raw Point Clouds , 2017, Comput. Graph. Forum.

[32]  Sunghee Choi,et al.  The power crust , 2001, SMA '01.

[33]  Zuoqiang Shi,et al.  Surface Reconstruction Based on the Modified Gauss Formula , 2018, ACM Trans. Graph..

[34]  Wei Wu,et al.  PointCNN: Convolution On X-Transformed Points , 2018, NeurIPS.