Heat Kernel Signature of 2D Shapes and its Application in Classification

被引：0

作者：

Sun D. ^{[1
,2
]}

Chen S. ^{[1
]}

Zhou Y. ^{[1
]}

Chen N. ^{[3
]}

Xin S. ^{[4
]}

Wang R. ^{[2
]}

机构：

[1] Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo

[2] College of Electronics and Computer, Zhejiang Wanli University, Ningbo

[3] Department of Computer Science, The University of Hong Kong, Hong Kong

[4] School of Computer and Science, Shandong University, Ji'nan

来源：

Chen, Shuangmin (chenshuangmin@nbu.edu.cn) | 2018年 / Institute of Computing Technology卷 / 30期

关键词：

3D modeling; Heat kernel feature; Numerical optimization; Shape classification;

D O I：

10.3724/SP.J.1089.2018.16780

中图分类号：

学科分类号：

摘要：

To seek for an isometry-invariant 2D shape descriptor, we encode 2D shapes from 3D perspective, and propose a novel shape classification approach based on heat kernel. First, we build triangulation for the region enclosed by the contour. Then, we transform the 2D shape into a 3D closed/smooth surface through a set of op-timization techniques. Finally, the heat kernel signature of the 3D counterpart is extracted to identify the original 2D shape. Extensive experimental results on the MPEG-7 and Animal Shapes benchmarks exhibit an advantage of classification in terms of accuracy and robustness. © 2018, Beijing China Science Journal Publishing Co. Ltd. All right reserved.

引用

页码：1431 / 1437

页数：6

共 25 条

[1] El-Ghazal A., Basir O., Belkasim S., Invariant curva-ture-based Fourier shape descriptors, Journal of Visual Communication and Image Representation, 23, 4, pp. 622-633, (2012)
[2] Belongie S., Malik J., Puzicha J., Shape matching and object recognition using shape contexts, IEEE Transactions on Pattern Analysis and Machine Intelligence, 24, 4, pp. 509-522, (2002)
[3] Aslan C., Erdem A., Erdem E., Et al., Disconnected skeleton: shape at its absolute scale, IEEE Transactions on Pattern Analysis and Machine Intelligence, 30, 12, pp. 2188-2203, (2008)
[4] Sun K.B., Super B.J., Classification of contour shapes using class segment sets, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2, pp. 727-733, (2005)
[5] Peter A.M., Rangarajan A., Maximum likelihood wavelet density estimation with applications to image and shape matching, IEEE Transactions on Image Processing, 17, 4, pp. 458-468, (2008)
[6] Shu X., Pan L., Wu X.J., Multi-scale contour flexibility shape signature for Fourier descriptor, Journal of Visual Communication and Image Representation, 26, pp. 161-167, (2015)
[7] Mokhtarian F., Mackworth A., Scale-based description and recognition of planar curves and two-dimensional shapes, IEEE Transactions on Pattern Analysis and Machine In-telligence, PAMI-8, 1, pp. 34-43, (1986)
[8] Ling H.B., Jacobs D.W., Shape classification using the in-ner-distance, IEEE Transactions on Pattern Analysis and Machine Intelligence, 29, 2, pp. 286-299, (2007)
[9] Wang J.W., Bai X., You X.G., Et al., Shape matching and classification using height functions, Pattern Recognition Letters, 33, 2, pp. 134-143, (2012)
[10] Chen S., Chen Y., Xin S., Et al., Topo-logical shortcut independent geodesic distance and its ap-plication, Journal of Computer-Aided Design & Com-Puter Graphics, 27, 11, pp. 2149-2154, (2015)

← 1 2 3 →