Pyramid U-Network for Skeleton Extraction from Shape Points

被引：13

作者：

Atienza, Rowel ^{[1
]}

机构：

[1] Univ Philippines, Elect & Elect Engn Inst, Quezon City, Philippines

来源：

2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2019) | 2019年

关键词：

RECOGNITION;

D O I：

10.1109/CVPRW.2019.00155

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The knowledge about the skeleton of a given geometric shape has many practical applications such as shape animation, shape comparison, shape recognition, and estimating structural strength. Skeleton extraction becomes a more challenging problem when the topology is represented in point cloud domain. In this paper, we present the network architecture, PSPU-SkelNet, for TeamPH which ranked 3rd in Point SkelNetOn 2019 challenge [2]. PSPU-SkelNet is a pyramid of three U-Nets that predicts the skeleton from a given shape point cloud. PSPU-SkelNet achieves a Chamfer Distance (CD) of 2.9105 on the final test dataset. The code of PSPU-SkelNet is available at https://github.com/roatienzalskelnet.

引用

页码：1177 / 1180

页数：4

共 16 条

[1]

[Anonymous], 2016, PROC CVPR IEEE, DOI DOI 10.1109/CVPR.2016.265

[2]

[Anonymous], 2017, IEEE P COMPUT VIS PA, DOI DOI 10.1109/CVPR.2017.16

[3] Analysis of two-dimensional non-rigid shapes [J].

Bronstein, Alexander M. ;

Bronstein, Michael M. ;

Bruckstein, Alfred M. ;

Kimmel, Ron .

INTERNATIONAL JOURNAL OF COMPUTER VISION, 2008, 78 (01) :67-88

[4]

Demir Ilke, 2019, SKELNETON 2019 DATAS

[5] The Propagated Skeleton: A Robust Detail-Preserving Approach [J].

Durix, Bastien ;

Chambon, Sylvie ;

Leonard, Kathryn ;

Mari, Jean-Luc ;

Morin, Geraldine .

DISCRETE GEOMETRY FOR COMPUTER IMAGERY, DGCI 2019, 2019, 11414 :343-354

[6]

Goodfellow IJ, 2014, ADV NEUR IN, V27, P2672

[7]

Isola P, 2017, PROC CVPR IEEE, P1125, DOI DOI 10.1109/CVPR.2017.632

[8]

Kingma DP, 2014, ARXIV

[9]

Leborgne A, 2014, LECT NOTES COMPUT SC, V8887, P293, DOI 10.1007/978-3-319-14249-4_28

[10]

Leonard K, 2016, INT C PATT RECOG, P3216, DOI 10.1109/ICPR.2016.7900130

← 1 2 →