基于表观的动态孤立手势识别

被引:16
作者
祝远新
徐光祐
黄浴
机构
[1] 清华大学计算机科学与技术系!北京
关键词
计算机视觉; 人机接口; 手势识别; 图像运动模型; 鲁棒回归; 动态规划匹配;
D O I
10.13328/j.cnki.jos.2000.01.006
中图分类号
TP391.4 [模式识别与装置];
学科分类号
0811 ; 081101 ; 081104 ; 1405 ;
摘要
给出一种基于表观的动态孤立手势识别技术 .借助于图像运动的变阶参数模型和鲁棒回归分析 ,提出一种基于运动分割的图像运动估计方法 .基于图像运动参数 ,构造了两种表观变化模型分别作为手势的表观特征 ,利用最大最小优化算法来创建手势参考模板 ,并利用基于模板的分类技术进行识别 .对 12 0个手势样本所做的大量实验表明 ,这种动态孤立手势识别技术具有识别率高、计算量小、算法稳定性好等优点 .
引用
收藏
页码:54 / 61
页数:8
相关论文
共 10 条
[1]  
On creating reference templates for speaker independent interpretation of isolated words. Rabiner L R. IEEE Transactions on Acoustics Speech and Signal Processing . 1978
[2]  
Task-Specific gesture analysis in real -time using interpolated views. Darrell T J, Essa I A, Pentland A P. IEEE Transactions on Pattern Analysis and Machine Intelligence . 1996
[3]  
智能接口与智能应用新进展——第 2届全国智能接口与智能应用学术会议论文集. 祝远新,徐光礻右,俞志和. 电子工业出版社 . 1997
[4]  
Dynamic programming optimization for spoken word interpretation. Sakoe H, Chiba S. IEEE Transactions on Acoustics,Speech, and Signal Processing . 1978
[5]  
Motion-Based segmentation scheme to feature extraction of hand gestures. Zhu Yuanxin,Huang Yu,Xu Guangyou et al. Proceedings of SPIE the International Society for Optical Engineering . 1998
[6]  
Visual interpretation of hand gestures for humancomputer interaction: A review. Pavlovic V,Sharma R,Huang T S. IEEE transactions on pattern analysis and machineintelligence . 1997
[7]  
The robust estimation of multiple motions: parametric and piecewise-smooth flow fields. Black M J, Anandan P. Computer Vision and Image Understanding . 1996
[8]  
Eyes in the interface. Quek F. Image and Vision Computing . 1995
[9]  
Towards 3D hand tracking using a deformable model. Heap T,Hogg D. Proceedings of the 2 nd Interna-tional Conference on Automatic Face and Gesture Recognition . 1996
[10]  
A threshold selection method from gray-level histogram. Otsu N. IEEE Transactions on Systems Man and Cybernetics . 1979