Self-Supervised Non-Isometric 3D Shape Collection Correspondence Calculation Method

被引：0
|
作者：
Wu Y. [1 ,2 ]
Yang J. [1 ,3 ]
Zhang S. [1 ]
机构：
[1] School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou
[2] School of Big Data and Artificial Intelligence, Fujian Polytechnic Normal University, Fuqing
[3] Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou
来源：
Moshi Shibie yu Rengong Zhineng/Pattern Recognition and Artificial Intelligence | 2024年 / 37卷 / 03期
基金：
中国国家自然科学基金;
关键词：
Correspondence; Deep Learning; Intrinsic-Extrinsic Feature Alignment; Non-Isometric Shape Collection; Self-Supervised;
D O I：
10.16451/j.cnki.issn1003-6059.202403006
中图分类号：
学科分类号：
摘要：
Aiming at the problem of low accuracy and poor generalization ability in existing non-isometric 3D shape collection correspondence calculation methods, a self-supervised non-isometric 3D shape collection correspondence calculation method using deep intrinsic-extrinsic feature alignment algorithm is proposed. Firstly, discriminative feature descriptors are obtained by directly learning the original 3D shape features through DiffusionNet. Then, the deep intrinsic-extrinsic feature alignment algorithm is employed to compute correspondences between non-isometric shapes. Consistency between internal and external information is realized by utilizing local manifold harmonic bases as intrinsic information of the shapes and integrating external information such as Cartesian coordinates. Consequently, correspondence results are generated automatically in an unsupervised manner. Finally, a weighted undirected graph of non-isometric shape collections is constructed. Based on the principle of inherent correlation among similar geometric shapes, a self-supervised multi-shape matching algorithm is designed to continuously enhance the cycle-consistency of the shortest path in the shape graph, and thus optimal correspondences for non-isometric 3D shape collections are obtained. Experimental results demonstrate that the proposed method achieves small geodesic errors in correspondences with accurate results, and effectively deals with the symmetric ambiguity problem with good generalization ability. © 2024 Science Press. All rights reserved.
引用
收藏
页码：253 / 266
页数：13
相关论文
共 38 条
  • [1] Kaick V.O., Zhang H., Hamarneh G., Et al., A Survey on Shape Correspondence, Computer Graphics Forum, 30, 6, pp. 1681-1707, (2011)
  • [2] Sahillioglu Y., Recent Advances in Shape Correspondence, The Visual Computer, 36, 8, pp. 1705-1721, (2020)
  • [3] Deng B.L., Yao Y.X., Dyke R.M., Et al., A Survey of Non-Rigid 3D Registration, Computer Graphics Forum, 41, 2, pp. 559-589, (2022)
  • [4] Cosmo L., Rodola E., Masci J., Et al., Matching Deformable Objects in Clutter, Proc of the 4th International Conference on 3d Vision, (2016)
  • [5] Ovsjanikov M., Ben-Chen M., Solomon J., Et al., Functional Maps: A Flexible Representation of Maps Between Shapes, ACM Transactions on Graphics, 31, 4, (2012)
  • [6] Melzi S., Rodola E., Castellani U., Et al., Localized Manifold Harmonics for Spectral Shape Analysis, Computer Graphics Forum, 37, 6, pp. 20-34, (2018)
  • [7] Mandad M., Cohen-Steiner D., Kobbelt L., Et al., Variance-Minimizing Transport Plans for Inter-Surface Mapping, ACM Transactions on Graphics, 36, 4, (2017)
  • [8] Ezuz D., Solomon J., Ben-Chen M., Reversible Harmonic Maps between Discrete Surfaces, ACM Transactions on Graphics, 38, 2, (2019)
  • [9] Edelstein M., Ezuz D., Ben-Chen M., ENIGMA: Evolutionary Non-Isometric Geometry Matching, ACM Transactions on Graphics, 39, 4, (2020)
  • [10] Eisenberger M., Lahner Z., Cremers D., Smooth Shells: Multi-scale Shape Registration with Functional Maps, Proc of the IEEE/ CVF Conference on Computer Vision and Pattern Recognition, pp. 12262-12271, (2020)
1234