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...

采用弱预测机制的人体运动跟踪算法

DOI: 10.11834/jig.200304150

Keywords: 计算机图象处理(520?6040),运动捕获,人体运动跟踪,多分辨率跟踪,特征对应

Full-Text   Cite this paper   Add to My Lib

Abstract:

快速运动和自遮挡是人体运动跟踪的难点所在.为此提出了一种采用弱预测机制的人体运动跟踪算法.该算法首先通过全局搜索,确定候选人体特征集;然后建立特征的色彩、运动等属性的时变模型,构造贝叶斯分类器,实现特征对应;最后根据人体特征层次模型,检验特征匹配,并实现被遮挡特征的定位.为提高跟踪效率,采用了基于图象多分辨率表示的特征搜索算法,由低分辨率图象通过全局搜索来获取初始候选特征集,然后在高分辨率下,不断改善候选特征精度.实验结果表明,该算法能实现对快速人体运动的跟踪并有效解决自遮挡问题.

References

[1]  [1]Bregler. C, Malik. J. Tracking people with twists and exponential maps [A]. In: Proceeding of IEEE Conference Computer Vision Pattern Recognition[C]. Santa Barbara, CA,USA: IEEE Computer Society Press, 1998:8~15.
[2]  [2]Fua P, Gruen A, Plankers R et al. Human body modeling and motion analysis from video sequences[J]. International Archives of Photogrammetry and Remote Sensing, 1998,32 (B5): 866 ~873.
[3]  [3]Intille S S, Bobick A F. closed world tracking [A]. In:Proceedings of the Fifth International Conference on Computer Vision[C]. Boston, MA,USA: IEEE Computer Society Press,1995: 672~678.
[4]  [5]Jepson A, Black M J. Mixture models for optical flow computation[A]. In: Proceedings of IEEE Conference Computer Vision Pattern Recognition [C]. New York: IEEE Computer Society Press, 1993: 760~ 761.
[5]  [7]Rohr K. Human movement analysis based on explicit motion models [A]. In: Motion based recognition [C]. Boston, MA,USA: Kluwer Academic Publishers, 1997:171 ~ 198.
[6]  [9]Grassia F Sebastian. Motion editing: Mathematical foundations[EB/OL]. http ://www. cs. cmu. edu/~ spiff/moedit9 9/mathfoundations. pdf, 2001-03-28.
[7]  [4]Rehg J, Kanade T. Model-based tracking of self-occluding articulated objects [A]. In: Proceedings of the Fifth International Conference on Computer Vision[C]. Boston, MA,USA: IEEE Computer Society Press, 1995: 612~617.
[8]  [6]Moeslund T B, Granum E. Multiple cues used in model-based human motion capture [A]. In: The Fourth International Conference on Automatic Face and Gesture Recognition [C].Grenoble, France: IEEE Computer Society Press, 2000:362~367.
[9]  [8]Pavlovi\'c V, Rehg J M, Chan T J et al. A dynamic bayesian network approach to figure tracking using learned dynamic models [A]. In: International Conference on Computer Vision[C]. Corfu, Greece: IEEE Computer Society Press, 1999: 94~101.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133