增强现实中虚实结合的新方法

摘要
图/表
参考文献
相关文章 (4)

全文: PDF (519 KB) HTML ( ) 输出: BibTeX \| EndNote (RIS)
引用本文:
全红艳, 王长波, 林俊隽. 增强现实中虚实结合的新方法[J]. 机器人, 2008, 30(6): 561-565.. QUAN Hong-yan, WANG Chang-bo, LIN Jun-juan. A New Method of Virtual and Real Object Correspondence in Augmented Reality. ROBOT, 2008, 30(6): 561-565..

摘要提出了一种增强现实中虚实结合的新方法．首先研究了一种两帧相邻图像的快速点对应方法；在准确对应的前提下，大大提高了对应的速度．在此基础之上，研究了基于平面结构的虚实结合方法；通过计算表示虚实坐标之间关系的单应矩阵，可以得到任意一帧虚实坐标之间的单应矩阵，实现了虚实结合和虚拟物体的动态跟踪．所提方法简单、方便，对于虚拟物体的注册误差较低．

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	全红艳
	王长波
	林俊隽

关键词 ：增强现实, 平面结构, 单应矩阵, 注册

Abstract：A new method is proposed for virtual and real object correspondence in augmented reality.A fast point corre- spondence method for two adjacent images is studied,which can greatly improve the correspondence accuracy and speed in the condition of correct correspondence.Then a virtual and real object correspondence method based on planar structure is studied.Homography matrix between virtual and real coordinates of any frame can be obtained after the homography matrix between virtual and real coordinates is calculated.With this method,the virtual and real objects can be combined and the virtual objects can be dynamically tracked.The presented method is simple and convenient,and has a lower error in virtual object registration.

Key words： augmented reality planar structure homography matrix registration

收稿日期: 2008-05-05

TP391

基金资助:国家青年自然科学基金资助项目（60603076）；上海市科委自然科学基金资助项目（07ZR14035）；国家863计划资助项目（2007AA01Z316）；上海市青年科技启明星计划资助项目（08QA14025）

作者简介: 全红艳（1968- ），女，副教授．研究领域：虚拟现实，增强现实，基于图像绘制和计算机视觉．
王长波（1976- ），男，副教授．研究领域：虚拟现实，增强现实，计算机图形学和计算机视觉．

链接本文:

https://robot.sia.cn/CN/ 或 https://robot.sia.cn/CN/Y2008/V30/I6/561

[1] Pressigout M,Marchand E.Hybrid tracking algodthms for planar and non-planar structures subject to illumination changes[A].Proceedings of the IEEE/ACM International Symposium on Mixed and Augmented Reality[C].Piscamway,NJ,USA:IEEE,2006.52～55.
[2] Lourakis M I A,Halkidis S T,Orphanoudakis S C,et al.Matching disparate views of planar surfaces using projective invariants[J].Image and Vision Computing,2000,18(9):673～683.
[3] Simon G,Fitzgibbon A W,Zisserman A.Markerless tracking using planar structures in the scene[A].Proceedings of the IEEE and ACM Intermational Symposium on Augmented Reality[C].Piscataway,NJ,USA:IEEE,2000.120～128.
[4] Simon G,Berger M O.Reconstructing while registering:A novel approach for markerless augmented reality[A].Proceedings of the International Mixed and Augmented Reality[C].Piscataway,NJ,USA:IEEE,2002.285～294.
[5] Simon G,Berger M O.Real time registration known or recovered multi-planar structures:Application to AR[DB/OL].http://www.bmva.ac.uk/bmvc/2002/papers/91/full_91.pdf,2000/2008.
[6] Yuan M L,Ong S K,Nee A Y C.Registration based onprojective reconstruction technique for augmented reality systems[J].IEEE Transactions on Visualization and Computer Graphics,2005,11(3):254～264.
[7] Lourakis M I A,Argyms A A.Vision-based camera motion recovery for augmented reality[A].Proceedings of Computer Graphics International Conference[C].Piscataway,NJ,USA:IEEE,2004.569～576.
[8] Zhang Z Y,Deriche R,Faugeras O,et al.Robust technique for matching two uncalibrated images through the recovery of the unknown epipolar geometry[J].Articial Intelligence,1995,78(1-2):87～119.
[9] 马颂德,张正友.计算机视觉--计算理论与算法基础[M].北京:科学出版社,1998.52～56.Ma Song-de,Zhang Zheng-you.Computer Vision:Fundamentals of Computational Theory and Algorithms[M].Beijing:Science Press.1998.52～56.
[10] Chert C S,Hsieh W T.Composition of 3D graphic objects and panoramas[DB/OL].http://www.iis.sinica.edu.tw/～song/paper/ICATCamera.pdf,2000/2008.
[11] Uematsu Y,Saito H.Vision-based registration for augmented reality with integration of arbitrary multiple planes[A].Proceedings of the 13th International Conference on Image Analysis and Processing(LNCS vol.3617)[C].Heidelberg,Germany:Springer-Verlag,2005.155～162.