Multi-Precision Clothing Modeling Method Based on Subgraph Convolutional Neural Network

doi:10.19678/j.issn.1000-3428.0064194

Abstract

Abstract: Exsiting clothing simulation methods combining machine learning are simulated on a single precision grid, resulting in unnecessary calculation in the area with small deformation.Therefore, a multi-precision clothing modeling method based on a subgraph Convolutional Neural Network(CNN) is proposed for the physical simulation of clothes.The average deformation degree of each area of clothing is calculated using Rayleigh entropy curvature or continuous probability distribution.The threshold value of the clothing grid is divided according to the average deformation degree, and a multi-precision clothing grid corresponding to the original grid is constructed.Combined with the structural model of the human body, a multi-precision clothing graph structure based on time-space is extracted from the multi-precision clothing grid.The subgraph CNN is used to sample the neighbor nodes for a given vertex, and the vertex feature data are updated by aggregating the features of the given vertex and neighbor nodes.The experimental results show that with the use of the proposed method, the cloth computational efficiency improved by 25.3% compared with that of the TailorNet methods.The proposed method can not only retain the wrinkles learned from the physical simulation but also has a more realistic simulation effect and improves the computational efficiency.

Key words: clothing modeling, machine learning, multi-precision grid, graph Convolutional Neural Network(CNN), subgraph training

摘要： 现有融合机器学习的服装仿真方法大多在单一精度网格上进行仿真，导致在变形较小的区域内进行不必要的计算。提出一种基于子图卷积神经网络的多精度服装建模方法。采用基于物理模拟的方法进行服装仿真，利用瑞利熵曲率计算服装各区域的平均变形度，依据平均变形度对服装网格阈值进行划分，构建与原始网格相对应的多精度服装网格。结合人体结构化模型，从多精度服装网格中提取基于时空的多精度服装图结构。在此基础上，利用子图卷积神经网络为给定顶点采样邻居节点，通过聚合给定顶点和邻居节点的特征，以更新顶点特征数据。实验结果表明，与TailorNet方法相比，该方法的布料计算效率提升25.3%，不仅保留了从物理模拟中学习的褶皱，而且具有更加真实的模拟效果，并提高了计算效率。

关键词: 服装建模, 机器学习, 多精度网格, 图卷积神经网络, 子图训练

CLC Number:

TP391

CHEN Zhixu, JIN Yanxia, LU Ye, YANG Jing, LIU Yabian, SHI Zhiru. Multi-Precision Clothing Modeling Method Based on Subgraph Convolutional Neural Network[J]. Computer Engineering, 2023, 49(4): 174-181.

陈治旭, 靳雁霞, 芦烨, 杨晶, 刘亚变, 史志儒. 基于子图卷积神经网络的多精度服装建模方法[J]. 计算机工程, 2023, 49(4): 174-181.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0064194

http://www.ecice06.com/EN/Y2023/V49/I4/174

Figures/Tables 13

References

[1] 朱淮冰, 金小刚, 冯结青, 等.布料动画模拟综述[J].计算机辅助设计与图形学学报, 2004, 16(5):613-618. ZHU H B, JIN X G, FENG J Q, et al.Survey on cloth animation[J].Journal of Computer Aided Design & Computer Graphics, 2004, 16(5):613-618.(in Chinese)
[2] 梁秀霞, 韩慧健, 张彩明.基于物理仿真的布料动画研究综述[J].计算机研究与发展, 2014, 51(1):31-40. LIANG X X, HAN H J, ZHANG C M.Survey of cloth animation based on physical simulation[J].Journal of Computer Research and Development, 2014, 51(1):31-40.(in Chinese)
[3] NARAIN R, SAMII A, O'BRIEN J F.Adaptive anisotropic remeshing for cloth simulation[J].ACM Transactions on Graphics, 2012, 31(6):152.
[4] HAN H J, WANG L Y, LIU K.One method of cloth simulation based on adaptive meshes[C]//Proceedings of the 12th International Conference on Computational Intelligence and Security.Washington D.C., USA:IEEE Press, 2016:483-486.
[5] 钟李涛, 侯进, 龚随, 等.基于低精度布料采样的多精度布料构建方法[J].计算机应用研究, 2019, 36(1):278-283. ZHONG L T, HOU J, GONG S, et al.Multi-resolution cloth construction based on low-resolution cloth sampling[J].Application Research of Computers, 2019, 36(1):278-283.(in Chinese)
[6] GUAN P, REISS L, HIRSHBERG D A, et al.DRAPE:dressing any person[J].ACM Transactions on Graphics, 2012, 31(4):35.
[7] DENG Y W, MAO T L, SHI M, et al.Cloth deformation prediction based on human motion[C]//Proceedings of International Conference on Virtual Reality and Visualization.Washington D.C., USA:IEEE Press, 2016:258-263.
[8] 石敏, 杨柳, 毛天露, 等.服装动画中人体运动与服装变形的相关性学习[J].计算机辅助设计与图形学学报, 2017, 29(10):1941-1951. SHI M, YANG L, MAO T L, et al.Study on the correlation between human motion and garment deformation in cloth animation[J].Journal of Computer-Aided Design and Computer Graphics, 2017, 29(10):1941-1951.(in Chinese)
[9] OH Y J, LEE T M, LEE I K.Hierarchical cloth simulation using deep neural networks[EB/OL].[2022-02-10].https://arxiv.org/pdf/1802.03168.pdf.
[10] SANTESTEBAN I, OTADUY M A, CASAS D.Learning-based animation of clothing for virtual try-on[J].Computer Graphics Forum, 2019, 38(2):355-366.
[11] GUNDOGDU E, CONSTANTIN V, SEIFODDINI A, et al.GarNet:a two-stream network for fast and accurate 3D cloth draping[C]//Proceedings of International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2019:8738-8747.
[12] PATEL C, LIAO Z, PONS-MOLL G.TailorNet:predicting clothing in 3D as a function of human pose, shape and garment style[C]//Proceedings of Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2020:7363-7373.
[13] JIN N, ZHU Y, GENG Z, et al.A pixel-based framework for data-driven clothing[J].Computer Graphics Forum, 2020, 39(8):135-144.
[14] SCARSELLI F, GORI M, TSOI A C, et al.The graph neural network model[J].IEEE Transactions on Neural Networks, 2009, 20(1):61-80.
[15] BRUNA J, ZAREMBA W, SZLAM A, et al.Spectral networks and locally connected networks on graphs[C]//Proceedings of International Conference on Learning Representations.New York, USA:[s.n.], 2014:1-10.
[16] DEFFERRARD M, BRESSON X, VANDERGHEYNST P.Convolutional neural networks on graphs with fast localized spectral filtering[C]//Proceedings of the 30th International Conference on Neural Information Processing Systems.New York, USA:ACM Press, 2016:3844-3852.
[17] KIPF T N, WELLING M.Semi-supervised classification with graph convolutional networks[EB/OL].[2022-02-10].https://arxiv.org/pdf/1609.02907.pdf.
[18] HAMILTON W L, YING R, LESKOVEC J.Inductive representation learning on large graphs[EB/OL].[2022-02-10].https://arxiv.org/pdf/1706.02216.pdf.
[19] PETER V, CUCURULL G, CASANOVA A, et al.Graph attention networks[EB/OL].[2022-02-10].https://arxiv.org/pdf/1710.10903.pdf.
[20] MEYER M, DESBRUN M, SCHRÖDER P, et al.Discrete Differential-Geometry Operators for Triangulated 2-Manifolds[M]//HEGE H C, POLTHIER K.Visualization and Mathematics III.Berlin, Germany:Springer, 2003:35-57.
[21] GUNDOGDU E, CONSTANTIN V, PARASHAR S, et al.GarNet:improving fast and accurate static 3D cloth draping by curvature loss[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(1):181-195.
[22] 王占东.细分曲面关键技术研究[D].南京:南京航空航天大学, 2003. WANG Z D.Research on critical technology of subdivision surfaces[D].Nanjing:Nanjing University of Aeronautics and Astronautics, 2003.(in Chinese)
[23] GARLAND M, HECKBERT P S.Surface simplification using quadric error metrics[C]//Proceedings of the 24th Annual Conference on Computer Graphics and Interactive Techniques.New York, USA:ACM Press, 1997:209-216.
[24] LOPER M, MAHMOOD N, ROMERO J, et al.SMPL:a skinned multi-person linear model[J].ACM Transactions on Graphics, 2015, 34(6):248.
[25] 石敏, 毛天露, 王兆其.基于实例数据分析的多精度网格布料动画[J].计算机学报, 2015, 38(6):1296-1306. SHI M, MAO T L, WANG Z Q.Multi-resolution cloth based on animation instances[J].Chinese Journal of Computers, 2015, 38(6):1296-1306.(in Chinese)

Please choose a citation manager

Content to export