纺织学报 ›› 2021, Vol. 42 ›› Issue (09): 144-149.doi: 10.13475/j.fzxb.20201200406

• 服装工程 • 上一篇    下一篇

基于人体动态特征的三维服装虚拟试穿技术

黎博文, 王萍(), 刘玉叶   

  1. 东华大学 信息科学与技术学院, 上海 201620
  • 收稿日期:2020-12-02 修回日期:2021-05-26 出版日期:2021-09-15 发布日期:2021-09-27
  • 通讯作者: 王萍
  • 作者简介:黎博文(1996—),男,硕士生。主要研究方向为三维服装虚拟试穿技术。

3-D virtual try-on technique based on dynamic feature of body postures

LI Bowen, WANG Ping(), LIU Yuye   

  1. College of Information Science and Technology, Donghua University, Shanghai 201620, China
  • Received:2020-12-02 Revised:2021-05-26 Published:2021-09-15 Online:2021-09-27
  • Contact: WANG Ping

摘要:

针对人体个性化姿态的动态感知以及服装模型对人体姿态动作的跟踪匹配精度低的问题,提出了一种基于人体动态特征的三维服装骨骼点分层模型,区分躯干(整体)与四肢(局部)的运动特征,并融合了基于Kinect体感设备的人体围度数据;设计了整体-局部分解算法,分层控制三维服装骨骼模型对人体姿态动作的高精度跟踪。经真人虚拟试穿的测试结果表明:新方法的相对跟踪误差降低约10%,明显改善了虚拟服装对人体姿态的实时跟踪精度,提高了服装模型与真人体型的匹配精度,支持裙子、裤子、上衣等服装类型的复杂姿态(抬腿、弓步等)试穿场景。

关键词: 三维虚拟试穿, 人体动态特征, 协同跟踪, 服装骨骼点分层模型, 整体-局部分解

Abstract:

Aiming at the dynamic perception of personalized postures and the tracking and matching of human body's posture by clothing modeling, this paper established the hierarchical 3-D clothing skeleton points model based on two types of the motion features of the human postures, distinguishing the trunk (global) from the limbs (local), and integrating the clothing model with the body circumferencial features extracted using Kinect camera. In parallel, the global-local singular value decomposition algorithm (GL-SVD) was designed to hierarchically control the deformation of 3-D clothing skeleton model cooperatively. The results from virtual try-on test system show that relative tracking error of the new method is reduced by about 10%, the real-time tracking accuracy and matching precision between the clothing model and the human body is obviously improved. This new technique supports the complex posture (such as leg lifts, lunges) trying-on scenarios of various clothing types such as skirts, pants, tops, and more.

Key words: 3-D virtual try-on, dynamic features of human body, cooperative tracking, hierarchical model of clothing skeleton points, global-local singular value decomposition

中图分类号: 

  • TS942.8

图1

Kinect人体骨骼点模型"

图2

服装骨骼点模型"

表1

个性化人体围度特征"

性别 胸围/cm 腰围/cm 臀围/cm 大腿围/cm
93 84 88 34
86 72 85 28

图3

基于GL-SVD算法的协同跟踪"

图4

投影成像示意图"

图5

四元数旋转示意图"

图6

虚拟试穿系统原理图"

表2

虚拟试穿场景"

场景编号 图序 视角 姿态 服装类型
1 图7(b) 正面 侧抬腿 抹胸裙
2 图7(c) 正面 前抬腿 抹胸裙
3 图8(b) 正面 左右弓步 七分裤
4 图8(c) 正面 前后弓步 七分裤
5 图9(b) 正面 T字型 短袖上衣
6 图9(c) 侧面 站立 短袖上衣

图7

抬腿姿态试穿效果"

图8

弓步姿态试穿效果"

图9

上身姿态试穿效果"

图10

文献中Avatar方法试穿效果"

图11

跟踪精度和实时性分析"

图12

跟踪精度对比"

[1] YU L, ZHONG Y, WANG X. Inpainting-based virtual try-on network for selective garment transfer[J]. IEEE Access, 2019(7):134125-134136.
[2] HSIEH C, CHEN C, CHOU C, et al. Fit-me: image-based virtual try-on with arbitrary poses[C]// 2019 IEEE International Conference on Image Processing. [s.l.]: IEEE, 2019:4694-4698.
[3] LIU T, XU H, ZHANG X. 3-D clothing transfer in virtual fitting based on UV mapping[C]// International Congress on Image and Signal Processing. Beijing:IEEE, 2018:1-5.
[4] 潘博, 钟跃崎. 基于二维图像的三维服装重建[J]. 纺织学报, 2020, 41(4):123-128.
PAN Bo, ZHONG Yueqi. Image-based three-dimensional garment reconstruction[J]. Journal of Textile Research, 2020, 41(4):123-128.
[5] 朱伟明, 谢琴, 彭卉. 男西服数字化智能化量身定制系统研发[J]. 纺织学报, 2017, 38(4):151-157.
ZHU Weiming, XIE Qin, PENG Hui. Research and development of digital intelligent customization system for men's suits[J]. Acta textile Sinica, 2017, 38(4):151-157.
[6] 薛静雅. 基于Kinect的虚拟试衣系统设计[D]. 呼和浩特:内蒙古大学, 2018:12-45.
XUE Jingya. Design of virtual fitting system based on Kinect[D]. Huhhot: Inner Mongolia University, 2018:12-45.
[7] YUAN M, KHAN I R, FARBIZ F, et al. A mixed reality virtual clothes try-on system[J]. IEEE Transactions on Multimedia, 2013, 15(8):1958-1968.
doi: 10.1109/TMM.2013.2280560
[8] YOUSEF K M A, MOHD B J, AL-OMARI M. Kinect-based virtual try-on system: a case study[C]// Jordan International Joint Conference on Electrical Engineering and Information Technology (JEEIT). Amman Jordan: IEEE, 2019:91-96.
[9] 李涛, 冯向阳. 基于加权随机森林算法的虚拟试衣型号推荐技术研究[J]. 智能计算机与应用, 2019, 9(4):116-124.
LI Tao, FENG Xiangyang. Research on virtual fitting model recommendation technology based on weighted random forest algorithm[J]. Intelligent Computer and Application, 2019, 9(4):116-124.
[10] 万燕, 陈林下. 基于Kinect的虚拟试衣相关技术的研究[J]. 智能计算机与应用, 2018, 8(4):63-68.
WAN Yan, CHEN Linxia. Research on virtual fitting technology based on Kinect[J]. Intelligent Computer and Application, 2018, 8(4):63-68.
[11] LI B, HAN C, BAI B. Hybrid approach for human posture recognition using anthropometry and BP neural network based on Kinect V2[J]. EURASIP Journal on Image and Video Processing, 2019, 8(2019):1-4.
[1] 潘博, 钟跃崎. 基于二维图像的三维服装重建[J]. 纺织学报, 2020, 41(04): 123-128.
[2] 徐平华, 冒海琳, 黄琴琴. 服饰场景视觉共性认知评价方法[J]. 纺织学报, 2019, 40(09): 192-196.
[3] 穆红 林清华. 基于视觉搜索的服装店铺陈列优化[J]. 纺织学报, 2015, 36(09): 125-128.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!