基于特征点识别的头部姿态计算

乔体洲; 戴树岭

doi:10.13700/j.bh.1001-5965.2013.0530

基于特征点识别的头部姿态计算

doi: 10.13700/j.bh.1001-5965.2013.0530

北京航空航天大学自动化科学与电气工程学院, 北京 100191

基金项目:

国家高技术研究发展计划资助项目（2009AA01Z333）

详细信息

作者简介:
乔体洲（1983-），男，河北张家口人，博士生，qiao.tizhou@asee.buaa.edu.cn.

中图分类号: TP391.7
计量
- 文章访问数: 1309
- HTML全文浏览量: 70
- PDF下载量: 592
- 被引次数: 0
出版历程
- 收稿日期: 2013-09-23
- 网络出版日期: 2014-08-20

Head pose estimation framework based on feature point detection

School of Automation Science and Electrical Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

摘要

摘要: 为了提升使用随机回归森林进行头部姿态分析的精度，提出了一种基于特征点识别分析头部姿态的计算框架.考虑到高误差投票的干扰，该计算框架以随机森林的特征点识别为基础从而避免异常投票干扰，将头部姿态计算问题转换为空间鼻尖特征点和朝向特征点的识别问题.在随机森林的训练中，决策函数使用了高斯曲率和平均曲率作为图形特征，根据微分熵的信息增益在随机生成的决策函数库中搜索最优化决策函数.在训练完成的随机回归森林的叶子节点中，通过分析保存的样本数据，可以得到目标特征点的高斯分布估计.根据实验测试结果，在适当的阈值设定的情况下，该方法可以实现较高的识别成功率，使用曲率后明显提高了识别精度，能够在一定程度上处理有遮挡的数据，并且该方法已经成功应用于虚拟座舱的实时头部姿态分析计算系统.
- 特征点探测 /
- 面部朝向估计 /
- 随机森林 /
- 决策树 /
- 虚拟座舱
Abstract: In order to improve the precision of head pose estimation with random regression forest, an analysis framework based on feature point recognition was proposed for head pose estimation. In view of invalid votes disturbance, this framework recognized head position point and direction point to avoid accepting abnormal voting. The decision function used depth value, normal vector, Gaussian curvature and mean curvature as image features. An approximate optimized decision function search was executed in a binary test pool generated randomly according to information gain of differential entropy. The experiments focused on performance analysis on different occlusion rates of original data. The approach got high success rate in experiments with appropriate parameters, improved accuracy after using curvature, and enhanced the capability of handling with occlusion. The proposed framework has been applied for real-time head pose estimation system in virtual cockpits successfully.
- feature point detection /
- head pose estimation /
- random forest /
- decision tree /
- virtual cockpit

HTML全文

参考文献(17)

[1]	Martins P, Batista J.Accurate single view model-based head pose estimation[C]//Proceedings of International Conference on Automatic Face and Gesture Recognition.Piscataway,NJ:IEEE Computer Society Press,2008:4813369
[2]	Morency L P, Whitehill J,Movellan J.Generalized adaptive view-based appearance model:integrated framework for monocular head pose estimation[C]//Proceedings of International Conference on Automatic Face and Gesture Recognition.Piscataway,NJ:IEEE Computer Society Press,2008:4813429
[3]	Murphy-Chutorian E, Trivedi M M.Head pose estimation in computer vision:a survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(4):607-626
[4]	Breitenstein M D, Kuettel D,Weise T,et al.Real-time face pose estimation from single range images[C]//Proceedings of International Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Computer Society Press,2008:4587807
[5]	Lu X G, Jain A K.Automatic feature extraction for multiview 3D face recognition[C]//Proceedings of International Conference on Automatic Face and Gesture Recognition.Piscataway,NJ:IEEE Computer Society Press,2006:585-590
[6]	Weise T, Leibe B,Van G L.Fast 3d scanning with automatic motion compensation[C]//Proceedings of International Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Computer Society Press,2007:4270316
[7]	Breitenstein M D, Jensen J,Hilund C,et al.Head pose estimation from passive stereo images[C]//Lecture Notes in Computer Science.Heidelberg:Springer-Verlag,2009:219-228
[8]	Breiman L. Random forests[J].Machine Learning,2001, 45(1): 5-32
[9]	Gall J, Lempitsky V.Class-specific hough forests for object detection[C]//Proceedings of International Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Computer Society Press,2009:1022-1029
[10]	Criminisi A, Shotton J,Robertson D,et al.Regression forests for efficient anatomy detection and localization in CT studies[C]//Lecture Notes in Computer Science.Heidelberg:Springer-Verlag,2010:106-117
[11]	Huang C, Ding X Q,Fang C.Head pose estimation based on random forests for multiclass classification[C]//Proceedings of International Conference on Pattern Recognition.Piscataway,NJ:IEEE Computer Society Press,2010:934-937
[12]	Fanelli G, Gall J,Van G L.Real time head pose estimation with random regression forests[C]//Proceedings of International Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE Computer Society Press,2011:617-624
[13]	Fanelli G, Weise T,Gall J,et al.Real time head pose estimation from consumer depth cameras[C]//Lecture Notes in Computer Science.Heidelberg:Springer-Verlag,2011:101-110
[14]	Tang Y Q, Sun Z N,Tan T N.Real-time head pose estimation using random regression forests[C]//Lecture Notes in Computer Science.Heidelberg:Springer-Verlag,2011:66-73
[15]	Meyer M, Desbrun M,Schröder P,et al.Discrete differential-geometry operators for triangulated 2-manifolds[J].Visualization and Mathematics,2002,3(2):52-58
[16]	Gall J, Yao A,Razavi N,et al.Hough forests for object detection, tracking, and action recognition[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(11):2188-2202
[17]	Paysan P, Knothe R,Amberg B,et al.A 3D face model for pose and illumination invariant face recognition[C]//Proceedings of International Conference on Advanced Video and Signal Based Surveillance.Piscataway,NJ:IEEE Computer Society Press,2009:296-301