首页 | 本学科首页   官方微博 | 高级检索  
     

基于三维模型与Gabor小波的人脸特征点跟踪方法
引用本文:战江涛,刘强,柴春雷.基于三维模型与Gabor小波的人脸特征点跟踪方法[J].浙江大学学报(自然科学版 ),2011,45(1):30-36.
作者姓名:战江涛  刘强  柴春雷
作者单位:1.浙江大学 计算机科学与技术学院,浙江 杭州 310027; 2.浙江机电职业技术学院 设计艺术系,浙江 杭州 310053; 3.清华大学 美术学院,北京 100084
基金项目:国家自然科学基金资助项目(60703041);中央高校基本科研业务费专项资金资助项目(2009QNA5014).
摘    要:为了实现人脸特征点跟踪系统的鲁棒性和精确性,使用光束平差法模型将基于三维人脸几何模型的跟踪方法与基于Gabor小波的特征点跟踪方法结合起来,提出基于三维模型与Gabor小波的人脸特征点跟踪方法.该方法使用基于Gabor小波的特征跟踪方法来得到每帧的初始特征点,使用基于三维人脸几何模型的整合优化跟踪方法来得到每帧的最终特征点.与3类典型人脸特征点跟踪方法的对比实验结果表明,该方法克服了以往方法基于二维图像信息来寻找特征点的局限性,可以实现鲁棒的、实时的、大角度的人脸特征点跟踪.

关 键 词:人脸特征  Gabor小波  三维跟踪  光束平差法

Facial feature tracking using three-dimensional model and Gabor wavelet
ZHAN Jiang-tao,LIU Qiang,CHAI Chun-lei.Facial feature tracking using three-dimensional model and Gabor wavelet[J].Journal of Zhejiang University(Engineering Science),2011,45(1):30-36.
Authors:ZHAN Jiang-tao  LIU Qiang  CHAI Chun-lei
Affiliation:1.College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China;2.Department of Design and Art, Zhejiang Institute of Mechanical and Electrical Engineering, Hangzhou 310053, China; 3. Academy of Arts and Design, Tsinghua University, Beijing 100084, China
Abstract:A new facial feature points tracking scheme was proposed by integrating three-dimensional geometric face model based tracking method and facial feature points tracking method with Gabor wavelet in bundle adjustment way in order to realize the robustness and accuracy of facial tracking system. The tracking method with Gabor wavelet was used to get the initial feature points. Then the model-based tracking method was used to get the final optimal feature points. The scheme overcame the disadvantage of using two-dimensional image features compared with three other typical methods. The scheme can be used for robust, real-time and wide-angle facial feature tracking.
Keywords:facial feature  Gabor wavelet  three-dimensional tracking  bundle adjustment
点击此处可从《浙江大学学报(自然科学版 )》浏览原始摘要信息
点击此处可从《浙江大学学报(自然科学版 )》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号