全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

一种基于光流场重建三维运动和结构的新方法

DOI: 10.11834/jig.200306223

Keywords: 计算机图象处理(520?6040),角点跟踪,位移场,光流场,三维运动和结构,线性方法

Full-Text   Cite this paper   Add to My Lib

Abstract:

提出了一种基于稀疏光流场计算三维运动和结构的线性新方法,该方法综合视觉运动分析中的两类处理方法,选取图象中的角点作为特征点;并检测和跟踪图象序列中的角点.记录检测到的角点在图象序列中的位移,在理论上证明了时变图象的光流场可以近似地用角点的位移场代替,从而得到时变图象的稀疏光流场;通过光流运动模型的建立,推导出由稀疏光流场重建三维物体运动和结构的线性方法.通过用真实图象序列验证该算法,表明该算法取得了较好的效果

References

[1]  [3]Shi J,Tomasi C. Good features to track[A]. In:IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR\'94) [C]. Seattle, Washington, USA, 1994: 593 ~ 600.
[2]  [5]Azarbayejani A, Pentland A. Recursive estimation of motion,structure, and focal length[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1995, 17(6): 562~575.
[3]  [6]Chiuso A, Soatto S. 3-D Motion and structure from 2-D motion causally integrated over time. Part I: theory [R]. Technical Report, ESSRL 99-03, Washington University 2000.
[4]  [7]Chiuso A, Favaro P, Jin H et al. 3D motion and structure causally integrated over time: Implementation[A]. In:European Conference on Computer Vision, ECCV2000 [C]. Dublin,Ireland, 2000: 734~750.
[5]  [1]王润生.图象理解[M].长沙:国防科技大学出版社,1994.
[6]  [2]Tomasi C, Kanade T. Detection and tracking of point features[R ]. Tech. Rept. CMU-CS-91132. Pittsburgh: CarnegieMellon University School of Computer Science. 1991.
[7]  [4]陈震,高满屯,沈允文.图象光流场计算技术研究进展[J].中国图象图形学报,2002,7A(5):434~439.
[8]  [8]Xiong Y, Shafer S. Dense structure from a dense optical flow sequence[J]. Computer Vision and Image Understanding, 1998,69(2): 222~245.
[9]  [9]Kanatani K. Geometric computation for machine vision [M].Oxford: Oxford Science Publications, 1993: 237~245.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133