logo

SCIENCE CHINA Information Sciences, Volume 61, Issue 5: 054101(2018) https://doi.org/10.1007/s11432-017-9331-9

3D shape co-segmentation via sparse and low rank representations

More info
  • ReceivedOct 30, 2017
  • AcceptedJan 17, 2018
  • PublishedApr 10, 2018

Abstract

In this paper, we propose a 3D shape co-segmentation method, which divides 3D shapes in the same category into consistent feature representations. We involve sparse and low-rank constraints to obtain compact feature representations among the 3D shapes. After pre-segmentation and feature extraction processes, we convert the co-segmentation problem into feature clustering issues. With the sparse and low-rank constraints, the initial geometry features are mapped into a compact coefficient space. Then, we gather the coefficients and weight them by a confidence weighting procedure. Finally, we apply fuzzy cuts method for optimization and achieve the final shape co-segmentation results. Experimental results on two public benchmarks demonstrate that our approach is robust for various 3D meshes, and outperforms other state-of-the-art approaches.


Acknowledgment

This work was supported by National Natural Science Foundation of China (Grant Nos. 61532003, 61421003).


Supplement

Videos and other supplemental documents.


References

[1] Hu R, Fan L, Liu L. Co-segmentation of 3D shapes via subspace clustering. Comput Graph Forum, 2012, 31: 1703-1713 CrossRef Google Scholar

[2] Shu Z, Qi C, Xin S. Unsupervised 3D shape segmentation and co-segmentation via deep learning. Comput Aided Geom Des, 2016, 43: 39-52 CrossRef Google Scholar

[3] van Kaick O, Xu K, Zhang H. Co-hierarchical analysis of shape structures. ACM Trans Graph, 2013, 32: 1 CrossRef Google Scholar

[4] Sidi O, van Kaick O, Kleiman Y. Unsupervised co-segmentation of a set of shapes via descriptor-space spectral clustering. ACM Trans Graph, 2011, 30: 1 CrossRef Google Scholar

[5] Huang Q, Koltun V, Guibas L. Joint shape segmentation with linear programming. ACM Trans Graph, 2011, 30: 1 CrossRef Google Scholar

[6] Guo K, Zou D, Chen X. 3D mesh labeling via deep convolutional neural networks. ACM Trans Graph, 2015, 35: 1-12 CrossRef Google Scholar

[7] Liu G, Lin Z, Yan S. Robust recovery of subspace structures by low-rank representation.. IEEE Trans Pattern Anal Mach Intell, 2013, 35: 171-184 CrossRef PubMed Google Scholar

[8] Tan H, Cheng B, Feng J. Low-n-rank tensor recovery based on multi-linear augmented Lagrange multiplier method. Neurocomputing, 2013, 119: 144-152 CrossRef Google Scholar

[9] Katz S, Tal A. Hierarchical mesh decomposition using fuzzy clustering and cuts. ACM Trans Graph, 2003, 22: 954-961 CrossRef Google Scholar

[10] Chen X, Golovinskiy A, Funkhouser T. A benchmark for 3D mesh segmentation. ACM Trans Graph, 2009, 28: 1 CrossRef Google Scholar

[11] Wang Y, Asafi S, van Kaick O. Active co-analysis of a set of shapes. ACM Trans Graph, 2012, 31: 1 CrossRef Google Scholar

  • Table 1   Co-segmentation accuracy comparison with Hu et al. on PSB and shape COSEG dataset
    Category Hu et al. Ours Category Hu et al. Ours
    Human 70.40 71.86 Plier 86.00 87.90
    Cup 97.40 97.96 Fish 85.60 87.34
    Glasses 98.30 98.60 Bird 71.50 78.77
    Airplane 83.30 86.58 Armadillo 87.30 88.52
    Ant 92.90 93.94 Vase 80.20 80.00
    Chair 89.60 90.83 Fourleg 88.70 88.98
    Octopus 97.50 97.65 Candelabra 93.90 94.07
    Table 99.00 99.37 Goblet 99.20 99.24
    Teddy 97.10 93.56 Guitar 98.00 98.40
    Hand 91.90 92.28 Lamp 90.70 92.07

Copyright 2019 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有

京ICP备18024590号-1