虚拟试衣中的曲面配准

李晓捷; 王小波; 宋丽梅

doi:10.3788/OPE.20152313.0534

您当前的位置：

首页 >

文章列表页 >

虚拟试衣中的曲面配准

更新时间：2020-08-13

- 虚拟试衣中的曲面配准
- Curvature surface registration in virtual try-on garment
- 光学精密工程 2015年23卷第10z期页码：533-538
- 作者机构：
  
  1. 天津工业大学艺术与服装学院天津,300387
  2. 天津工业大学电气工程与自动化学院天津,300387
- 作者简介：
- 基金信息：
  
  天津市高等学校科技发展计划基金资助项目(No.20130324)();国家自然科学基金面上项目(No.61078041)()
- DOI：10.3788/OPE.20152313.0534
  中图分类号： TP391.3
- 收稿日期：2015-06-02，
  
  修回日期：2015-06-23，
  
  纸质出版日期：2015-11-14
- 稿件说明：
移动端阅览
李晓捷, 王小波, 宋丽梅. 虚拟试衣中的曲面配准[J]. 光学精密工程, 2015,23(10z): 533-538

Li Xiao-jie, Wang Xiao-bo, Song Li-mei. Curvature surface registration in virtual try-on garment[J]. Editorial Office of Optics and Precision Engineering, 2015,23(10z): 533-538
李晓捷, 王小波, 宋丽梅. 虚拟试衣中的曲面配准[J]. 光学精密工程, 2015,23(10z): 533-538 DOI： 10.3788/OPE.20152313.0534.

Li Xiao-jie, Wang Xiao-bo, Song Li-mei. Curvature surface registration in virtual try-on garment[J]. Editorial Office of Optics and Precision Engineering, 2015,23(10z): 533-538 DOI： 10.3788/OPE.20152313.0534.

摘要

研究了三维光学测量法在人体测量及虚拟试衣曲面配准中的应用。利用双目视觉系统获得1:1的三维人体数据

然后对原始三维人体数据进行数据处理

获得精确人体模型

最后使用已有的服装数据进行虚拟试衣。为了减化试衣方法

将人体模型及服装模型视为刚体模型

根据两模型的高斯曲率及平均曲率极值点改进算法获得人体和服装的关键特征点

并将特征点对应集作为粗匹配矩阵。使用非线性最小距离优化算法——Levenberg-Marquardt算法

把模型关键特征点的最小距离作为误差参数

从而得到特征对应点集的优化变换矩阵

以减小实验配准误差。实验证明

Levenberg-Marquardt优化算法对于曲面配准有效

误差设为10 mm时

迭代次数为33时收敛速度较好。该方法简单、有效、稳定

对三维服装的计算机辅助设计、个性化虚拟试衣的发展具有一定的促进作用。

Abstract

This paper focuses on applications of 3D measurement to the human body measuring and curvature surface registration of virtual trying on garment. In which

the binocular vision was used to get 3D human body data of 1:1 scale; then the 3D human data were processed to obtain a precise human model; finally

the garment data were taken to implement the trying on garment. To simplify the try-on method

the human model and the garment model were treated as a rigid

and the extremum point of the Gaussian curvature and the mean curvature of the two models as a matrix of coarse matching. To reduce the registration error

nonlinear minimum distance optimization algorithm

Levenberg-Marquardt algorithm

was used

and the model of key feature points of minimum distance was taken as error parameters to obtain optimal transform matrix of corresponding feature point sets. Experimental results verify that Levenberg-Marquardt optimization algorithm for surface registration is effective

where the error setting is 10 mm and the iteration number is 33 times. It is a simple

effective and stable fitting algorithm for promoting the research of 3D garment Compute Aided Design(CAD) and personalized virtual try-on.

关键词

Keywords

references

曲学军,张璐. 基于双目视觉的三维测量方法[J]. 计算机仿真,2011,28(2):373-377. QU X J,ZHANG L. 3D measurement method based on binocular vision technique[J]. Computer Simulation,2011,28(2):373-377.(in Chinese)

刘晓利,彭翔,殷永凯,等. 大尺寸薄壳物体表面的三维光学自动检测[J]. 光学学报, 2011,31(3):0312006. LIU X L,PENG X,YIN Y K,et al..3D Auto-inspection for large thin-wall object[J]. ACTA OPTICA SINICA,2011,31(3):0312006.(in Chinese)

王欣,张明明,于晓,等. 应用改进迭代最近点方法的点云数据配准[J]. 光学精密工程,2012,20(9):2068-2077. WANG X, ZHANG M M,YU X, et al.. Point cloud registration based on improved iterative closest point method[J]. Opt. Precision Eng.,2012,20(9):2068-2077.(in Chinese)

王道档, 王福民, 陈茜茜,等. 基于Levenbery-Marquardt算法的点衍射三维坐标测量方法[J]. 光学学报,2014,34(8):0812001. WANG D D,WANG F M,CHEN Q Q, et al.. Three-dimensional coordinate measurement with point-diffraction interferometer based on Levenbery-Marquardt algorithm[J]. ACTA OPTICA SINICA,2014,34(8):0812001.(in Chinese)

ZHANG S. Digital multiple wavelength phase shifting algorithm[C]. Proceedings of SPIE-The International Society for Optical Engineering, 2009,7432, 74320N-1-74320N-11.

SONG L M, DONG X X, XI J T, et al.. A new phase unwrapping algorithm based on three wavelength phase shift profilometry method[J]. Optics & Laser Technology, 2013(45):319-329.

MEYER M,DEBUNNE G,OESBRUN M,et al..Interactive. Interactive animation of cloth-like objects for virtual reality[J]. Journal of Visualization and Computer Animation,2001,12(1):1-12.

洪梅. 点云数据特征提取算法的改进[D].东华理工大学,2013. HONG M. Algorithm Improvement of Feature Extraction of Point Cloud Data[D]. East China Institute of Technology,2013.(in Chinese)

张雨禾,耿国华,魏潇然. 散乱点云谷脊特征提取[J]. 光学精密工程,2015,23(1):310-318. ZHANG Y X,GENG G X,WEI X R. Valley-ridge feature extraction from point clouds[J].Opt. Precision Eng.,2015,23(1):310-318.(in Chinese)

LEVENBERG K. A method for the solution of certain non-linear problems in least quares[J]. The Quarterly of Applied Mathematics,1944,2:164-168.(in Chinese)

伏燕军,杨坤涛,邹文栋,等. 基于Levenberg-Marquardt算法的图像拼接[J]. 激光杂志, 2007,28(5):46-48. FU Y J,YANG K T, ZOU W D, et al.. Image mosaic based on levenberg-marquardt algorithm[J]. LASER JOURNAL,2007,28(5):46-48.(in Chinese)

浏览量

246

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据