利用网格卷积特征的三维形变目标分类

doi:10.3969/j.issn.0372-2112.2020.04.005

电子学报 ›› 2020, Vol. 48 ›› Issue (4): 648-653.DOI: 10.3969/j.issn.0372-2112.2020.04.005

利用网格卷积特征的三维形变目标分类

史聪伟¹, 赵杰煜^1,2, 陈瑜¹

1. 宁波大学信息科学与工程学院, 浙江宁波 315211;
2. 浙江省移动网络应用技术重点实验室, 浙江宁波 315211

收稿日期:2019-02-25 修回日期:2019-07-21 出版日期:2020-04-25
- 通讯作者:
- 赵杰煜
- 作者简介:
- 史聪伟 男,1992年出生,浙江宁波人.宁波大学信息科学与工程学院硕士研究生,主要从事三维图形处理、模式识别相关研究.E-mail:scw_vv@126.com;陈瑜女,1992年出生,河南安阳人.现为宁波大学信息科学与工程学院博士研究生,主要从事图形图像、深度学习等相关研究E-mail:chenyu_cycy@126.com
- 基金资助:
- 国家自然科学基金 (No.61571247）; 浙江省自然科学基金重点项目 (No.LZ16F030001）

3D Non-rigid Object Classification with Mesh Convolution Features

SHI Cong-wei¹, ZHAO Jie-yu^1,2, CHEN yu¹

1. Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, Zhejiang 315211, China;
2. Mobile Network Application Technology Key Laboratory of Zhejiang Province, Ningbo, Zhejiang 315211, China

Received:2019-02-25 Revised:2019-07-21 Online:2020-04-25 Published:2020-04-25
- Corresponding author:
- ZHAO Jie-yu
- Supported by:
- National Natural Science Foundation of China (No.61571247); Key program of National Natural Science Foundation of Zhejiang Province, China (No.LZ16F030001)

摘要/Abstract

摘要： 三维目标的形状变化给目标识别带来很大挑战，同时三维网格模型的不规则数据结构难以直接应用卷积运算提取三维目标特征.对此，本文提出了一种高效的三维形变目标的网格卷积特征表示方法，准确提取形状信息并进行分类.首先通过网格卷积运算获得形变目标中典型局部曲面形状分布，其次通过马尔科夫链对曲面形状的空间共现关系建模，从而形成三维模型的全局特征描述，最后采用支持向量机实现形变目标分类.该方法将连续多项式函数作为卷积模板，实现针对不规则数据结构的网格卷积运算，并且给出了卷积模板参数的无监督学习方法.在标准非刚性三维模型数据集SHREC10与SHREC15上的实验结果表明本文方法能有效提取三维网格模型的形状信息，分类准确率分别达到了92.88%与96.54%.

关键词: 三维形变模型, 网格卷积, 三维形状特征, 支持向量机

Abstract: 3D object recognition with shape changes is a challenging task. The irregular data structure of the mesh model prevents the operation of the conventional convolution, which brings difficulties to feature extraction of the 3D non-rigid objects. In this paper, we propose a method of mesh convolution for 3D non-rigid objects to extract shape features and use them for classification. Firstly, we obtain the distribution of typical patch shapes by the mesh convolution. Then, we model the spatial co-occurrence relationship by Markov chains to complete the global feature description. Finally, we use the support vector machine to classify the 3D objects. Our method adopts the continuous polynomial function as the convolution kernel for the irregular data structure, and learn the kernel by an unsupervised learning method. Experimental results on the standard non-rigid 3D model datasets show our method can effectively extract the features and achieve classification accuracy of 92.88% on SHREC10 and 96.54% on SHREC15,respectively.

Key words: non-rigid 3D model, mesh convolution, 3D shape feature, support vector machine

中图分类号:

TP391

史聪伟, 赵杰煜, 陈瑜. 利用网格卷积特征的三维形变目标分类[J]. 电子学报, 2020, 48(4): 648-653.

SHI Cong-wei, ZHAO Jie-yu, CHEN yu . 3D Non-rigid Object Classification with Mesh Convolution Features[J]. Acta Electronica Sinica, 2020, 48(4): 648-653.

参考文献

[1] GU Jiu-xiang,WANG Zhen-hua,JASON K,et al.Recent advances in convolutional neural networks[J].Pattern Recognition,2018,77:354-377.
[2] BRONSTEIN M M,BRUNA J,LECUN Y,et al.Geometric deep learning:going beyond Euclidean data[J].IEEE Signal Processing Magazine,2017,34(4):18-42.
[3] WU Zhi-rong,SONG Shu-ran,KHOSLA A,et al.3D shapeNets:A deep representation for volumetric shapes[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].NJ:IEEE,2015.1912-1920.
[4] 李海生,孙莉,武玉娟,等.非刚性三维模型检索特征提取技术研究[J].软件学报,2018,29(2):483-505. LI Hai-sheng,SUN Li,WU Yu-juan,et al.Survey on feature extraction techniques for non-rigid 3D shape retrieval[J].Journal of Software,2018,29(2):483-505.(in Chinese)
[5] MAHMOUDI M,SAPIRO G.Three-dimensional point cloud recognition via distributions of geometric distances[J].Graphical Models,2009,71(1):22-31.
[6] SIPIRAN I,BUSTOS B.Harris 3D:a robust extension of the Harris operator for interest point detection on 3D meshes[J].The Visual Computer,2011,27(11):963-976.
[7] REUTER M,WOLTER F E,PEINECKE N.Laplace-Beltrami spectra as ‘Shape-DNA’ of surfaces and solids[J].Computer-Aided Design,2006,38(4):342-366.
[8] RUSTAMOV R M.Laplace-Beltrami eigenfunctions for deformation invariant shape representation[A].Proceedings of the 5th Eurographics Symposium on Geometry Processing[C].Switzerland:Eurographics Association,2007.225-233.
[9] SUN J,OVSJANIKOV M,GUIBAS L J,et al.A concise and provably informative multi-scale signature based on heat diffusion[A].Proceedings of the Symposium on Geometry Processing[C].Switzerland:Eurographics Association,2009.1383-1392.
[10] BRONSTEIN M M,KOKKINOS I.Scale-invariant heat kernel signatures for non-rigid shape recognition[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].NJ:IEEE,2010.1704-1711.
[11] AUBRY M,SCHLICKEWEI U,CREMERS D,et al.The wave kernel signature:A quantum mechanical approach to shape analysis[A].IEEE International Conference on Computer Vision[C].NJ:IEEE,2011.1626-1633.
[12] HAN Zhi-zhong,LIU Zhen-bao,VONG Chi-man,et al.BoSCC:Bag of spatial context correlations for spatially enhanced 3D shape representation[J].IEEE Trans on Image Processing,2017,26(8):3707-3720.
[13] LI Jian-guo,WU Wei-xin,WANG Tao,et al.One step beyond histograms:Image representation using Markov stationary features[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].NJ:IEEE,2008.1-8.
[14] NI Bing-bing,YAN Shui-cheng,KASSIM A A,et al.Contextualizing histogram[A].Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition[C].NJ:IEEE,2009.1682-1689.
[15] GAO Z,YU Z,PANG X.A compact shape descriptor for triangular surface meshes[J].Computer-Aided Design,2014,53:62-69.

利用网格卷积特征的三维形变目标分类

3D Non-rigid Object Classification with Mesh Convolution Features

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	曹源, 宋迪, 胡小溪, 孙永奎. 基于改进时域多尺度散布熵与支持向量机的转辙机故障诊断[J]. 电子学报, 2023, 51(1): 117-127.
[2]	吴国庆, 王罗胜斌, 庞晨, 李永祯, 王雪松. 雷达极化域变焦角反组合体对抗方法：抗冲淡式干扰[J]. 电子学报, 2022, 50(12): 2969-2983.
[3]	孙晨峰, 吕卫民, 戴洪德, 张浩晨. 一种基于TimeGAN和OCSVM的多元退化设备小子样数据增广方法[J]. 电子学报, 2022, 50(11): 2678-2687.
[4]	杨国为, 万鸣华, 赖志辉, 张凡龙, 黄伟强. 具有合适拒识机制的高正确识别率分类器设计[J]. 电子学报, 2021, 49(8): 1569-1576.
[5]	罗忠涛, 夏杭, 卢琨, 何子述. 超视距雷达中距离-多普勒图的瞬态干扰自动识别方法[J]. 电子学报, 2021, 49(7): 1279-1285.
[6]	钱忠胜, 俞情媛, 宋涛, 朱懿敏, 祝洁, 赵畅. 基于支持向量机回归模型的测试用例生成与重用[J]. 电子学报, 2021, 49(7): 1386-1391.
[7]	张芳鑫, 王法松, 李睿, 左婷. GSSK‑VLC系统中基于SVM的LED选择算法[J]. 电子学报, 2021, 49(7): 1400-1405.
[8]	刘芳, 韩笑. 基于小波变换和深度网络的着陆地貌图像分类[J]. 电子学报, 2021, 49(11): 2171-2176.
[9]	张学军, 景鹏, 何涛, 孙知信. 基于变分模态分解的癫痫脑电信号分类方法[J]. 电子学报, 2020, 48(12): 2469-2475.
[10]	方佳艳, 刘峤. 具有同步化特征选择的迭代紧凑非平行支持向量聚类算法[J]. 电子学报, 2020, 48(1): 44-58.
[11]	周国华, 卢剑炜, 顾晓清, 殷新春. 基于简约凸壳的一类模糊支持向量机[J]. 电子学报, 2019, 47(8): 1708-1716.
[12]	郑威迪, 李志刚, 贾涵中, 高闯. 基于改进型鲸鱼优化算法和最小二乘支持向量机的炼钢终点预测模型研究[J]. 电子学报, 2019, 47(3): 700-706.
[13]	刘颖, 胡丹, 范九伦, 王富平, 李大湘. 基于融合特征的现勘图像检索结果优化算法[J]. 电子学报, 2019, 47(2): 296-301.
[14]	李凯, 李慧. 基于pinball损失的结构模糊孪生支持向量机[J]. 电子学报, 2019, 47(10): 2221-2227.
[15]	周文, 贾金原. 一种SVM学习框架下的Web3D轻量级模型检索算法[J]. 电子学报, 2019, 47(1): 92-99.