Abstract
Point cloud has become increasingly prevalent in 3D computer vision tasks owing to its simplicity, flexibility and powerful representation capability. However, for 3D shapes classification, how to obtain discriminative and effective feature descriptors from point clouds is still a challenging task. In this paper, we implement a point convolutional neural network combining multi-geometric feature to classify 3D shapes. We firstly obtain hidden geometric information from raw input point. And then a set of point information that consists of three-dimensional coordinates, normal vectors and curvatures is input to network. The extracted feature combining local geometric information can represent 3D shapes more effectively. Experiments on the Modelnet40 dataset show the effectiveness of the proposed method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Tangelder JWH, Veltkamp RC (2004) A survey of content based 3D shape retrieval methods. In: 6th international conference on shape modeling and applications. IEEE, LA, pp 145–156
Xie J, Fang Y, Zhu F et al (2015) Deepshape: deep learned shape descriptor for 3D shape matching and retrieval. In: 28th IEEE conference on computer vision and pattern recognition. IEEE, NY, pp 1275–1283 (2015)
Cıbuk M, Budak U, Guo Y et al (2019) Efficient deep features selections and classification for flower species recognition. Measurement 137:7–13
Vishwanathan SVM, Murty MN (2002) SSVM: a simple SVM algorithm. In: International joint conference on neural networks. IEEE, NY, pp 2393–2398 (2002)
Fang Y, Xie J, Dai G et al (2015) 3D deep shape descriptor. In: 28th IEEE conference on computer vision and pattern recognition. IEEE, NY, pp 2319–2328
Dai G, Xie J, Zhu F et al (2016) Learning a discriminative deformation-invariant 3D shape descriptor via many-to-one encoder. Pattern Recogn Lett 83:330–338
Su H, Maji S, Kalogerakis E et al (2015) Multi-view convolutional neural networks for 3D shape recognition. In: IEEE international conference on computer vision. IEEE, NY, pp 945–953
Li Y, Bu R, Sun M, et al (2018) PointCNN: Convolution on X-transformed points. In: 32nd conference on neural information processing systems. NIPS, LA, pp 820–830
Guo Y, Sohel F, Bennamoun M et al (2013) Rotational projection statistics for 3D local surface description and object recognition. Int J Comput Vision 105(1):63–86
Tombari F, Salti S, Di Stefano L (2010) Unique signatures of histograms for local surface description. In: Daniilidis K, Maragos P, Paragios N (eds) 11th european conference on computer vision. LNCS, vol 6313. Springer, Heidelberg, pp 356–369
Wohlkinger W, Vincze M (2011) Ensemble of shape functions for 3D object classification, In: The IEEE international conference on robotics and biomimetics. IEEE, NY, pp 2987–2992
Aldoma A, Tombari F, Rusu RB et al (2012) OUR-CVFH–oriented, unique and repeatable clustered viewpoint feature histogram for object recognition and 6DOF pose estimation. In: joint DAGM (German association for pattern recognition) and OAGM symposium. Springer, Heidelberg, pp 113–122
Maturana D, Scherer S (2015) VoxNet: a 3D convolutional neural network for real-time object recognition. In: IEEE/RSJ international conference on intelligent robots & systems. IEEE, NY, pp 922–928
Wang PS, Liu Y, Guo YX et al (2017) O-CNN: octree-based convolutional neural networks for 3D shape analysis. ACM Trans Graph (TOG) 36(4):72
Qi CR, Su, H (2017) Pointnet: deep learning on point sets for 3D classification and segmentation. In: 30th IEEE conference on computer vision and pattern recognition. IEEE, NY, pp 652–660
Klokov R, Lempitsky V (2017) Escape from cells: deep kd-networks for the recognition of 3D point cloud models. In: 16th IEEE international conference on computer vision. IEEE, NY, pp 863–872
Qi CR, Yi L, Su H, et al (2017) Pointnet++: deep hierarchical feature learning on point sets in a metric space. In: 31st conference on neural information processing systems. NIPS, LA, pp 5099–5108
Mitra NJ, Nguyen A.: Estimating surface normals in noisy point cloud data. In: 19th ACM symposium proceedings on computational geometry proceedings. World Scientific Publ Co. Pte. Ltd., Singapore, pp 322–328
Yang P, Qian X (2007) Direct computing of surface curvatures for point-set surfaces. Symp Point Based Graph 7:29–36
Amenta N, Kil YJ (2004) Defining point-set surfaces. ACM Trans Graph (TOG) 23(3):264–270
Dey TK, Sun J.: An adaptive MLS surface for reconstruction with guarantees. In: Third eurographics symposium on geometry processing. Eurographics Association, Aire-la-Ville, pp 43–52 (2005)
Wu Z, Song S, Khosla A, et al. (2015) 3D shapenets: a deep representation for volumetric shapes. In: 28th IEEE conference on computer vision and pattern recognition. IEEE, NY, pp 1912–1920
Acknowledgments
This work is partially supported by National Natural Science Foundation of China (61877002, 61702020), Beijing Natural Science Foundation grant number (4172013), Beijing Natural Science Foundation-Haidian Primitive Innovation Joint Fund grant number (L182007) and Beijing Municipal Commission of Education PXM2019-014213-000007.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zeng, G., Wu, Y., Li, H., Tan, L. (2020). 3D Shape Classification Based on Point Convolutional Neural Network Combining Multi-geometric Features. In: Jia, Y., Du, J., Zhang, W. (eds) Proceedings of 2019 Chinese Intelligent Systems Conference. CISC 2019. Lecture Notes in Electrical Engineering, vol 594. Springer, Singapore. https://doi.org/10.1007/978-981-32-9698-5_49
Download citation
DOI: https://doi.org/10.1007/978-981-32-9698-5_49
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-32-9697-8
Online ISBN: 978-981-32-9698-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)