纹理约束下的人脸特征点跟踪.pdfVIP

1000-9825/2004/15(11)1607 ©2004 Journal of Software 软 件 学 报 Vol.15, No.11 ∗ 纹理约束下的人脸特征点跟踪 1,2 1,2+ 1,2 宋 刚 , 艾海舟 , 徐光祐 1(清华大学 智能技术与系统国家重点实验室,北京 100084) 2(清华大学 计算机科学与技术系,北京 100084) Texture Constrained Facial Feature Point Tracking SONG Gang1,2, AI Hai-Zhou1,2+, XU Guang-You 1,2 1(State Key Laboratory of Intelligent Technology and Systems, Tsinghua University, Beijing 100084, China) 2(Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China) + Corresponding author: Phn: +86-10 E-mail: ahz@, /~ahz Received 2003-08-11; Accepted 2004-06-10 Song G, Ai HZ, Xu GY. Texture constrained facial feature point tracking. Journal of Software, 2004,15(11): 1607~1615. /1000-9825/15/1607.htm Abstract: In this paper, a facial feature point tracking scheme is proposed by integrating Lucas-Kanade optical flow tracking algorithm and the face alignment statistical model, DAM (direct appearance model), together in a Bayesian framework. The prediction of feature positions from Lucas-Kanade algorithm exploits the inter-frame correlations and accelerates the tracking speed. The texture-shape constraint under DAM improves the localization accuracy and robustness. Experiments show that this method adapts well to the various face movements. It can be used in face recognition or 3D face modeling. Key words: optical flow; feature point localization; feature point tracking; Bayesian posterior probability estimation 摘 要: 将 Lucas-Kanade 光流跟踪算法与人脸特征点定位的统计模型 DAM(direct appearance model)在 Bayesian 框架下结合起来,提出了视频中人脸特征点定位与跟踪的一种混合模型方法.利用 Lucas-Kanade 算法 预测人脸特征点的位置,充分利用了帧间的相关信息,提高了跟踪的速度.通过 DAM 中纹理对