《计算机应用研究》|Application Research of Computers

基于合成的通用头部网格模型的表情合成

Facial expression synthesis based on synthetic general head model

免费全文下载 (已被下载 次)  
获取PDF全文
作者 黄军伟,侯进,雷腾
机构 1.西南交通大学 信息科学与技术学院,成都 610031;2.南京大学 计算机软件新技术国家重点实验室,南京 210093
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2015)04-1240-04
DOI 10.3969/j.issn.1001-3695.2015.04.067
摘要 针对当前通用脸部网格模型的可操控性和逼真性难以同时兼顾的问题,提出了一种新的通用网格模型合成方法,该方法基于Candide-3模型和从建模软件中导出来的模型进行简化得到的模型进行结合来合成通用头部模型。在此模型的基础上合成出六大基本表情,对输入人脸正面照片使用ASM(active shape model)算法自动提取出特征点,利用特征点和径向基函数来调整合成的通用模型得到特定的表情。实验结果显示,该通用网格模型具有良好的操控性,且能完整表示整个头部结构,还能表达六大基本表情和特定表情。
关键词 Candide-3模型;简化模型;通用头部网格模型;表情合成
基金项目 国家自然科学基金资助项目(61371165)
浙江大学CAD&CG国家重点实验室开放课题(A1416)
计算机软件新技术国家重点实验室开放课题基金资助项目(KFKT2013B22)
四川省动漫研究中心2012年度科研项目(DM201204)
本文URL http://www.arocmag.com/article/01-2015-04-067.html
英文标题 Facial expression synthesis based on synthetic general head model
作者英文名 HUANG Jun-wei, HOU Jin, LEI Teng
机构英文名 1. School of Information Science & Technology, Southwest Jiaotong University, Chengdu 610031, China; 2. State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210093, China
英文摘要 Considering the problem that maneuverability and reality based on the current generic facial mesh model can’t balance, this paper proposed a new method that combined the Candide-3 model with simplified model exported from software to synthesize general head model. The method, firstly based on the model synthesized the six basic facial expressions, then extracted the feature point from the inputted face images using the ASM algorithm automatically and lastly obtained specific expressions through adjusting the synthetic general model using feature points and radial basis function. The experimental results show that the synthetic general head model has good maneuverability and can present the entire head structure and can express the basic expressions and specific expression.
英文关键词 Candide-3 model; simplified model; general head model; expression synthesis
参考文献 查看稿件参考文献
  [1] LEO M J, MANIMEGALAI D. 3D modeling of human face:a survey[C] //Proc of International Conference on Trends in Information Sciences and Computing. 2011:40-45.
[2] WANG Shu-fan, LAI Shang-hong. Reconstructing 3D face model with associated expression deformation from a single face image via constructing a low-dimensional expression deformation manifold[J] . IEEE Trans on Pattern Analysis and Machine Intelligence, 2011, 33(10):2116-2120.
[3] PATEL N, ZAVERI M. 3D model reconstruction and animation from single view face image[C] //Proc of International Conference on Audio, Language and Image Processing. 2012:674-682.
[4] BAEK S Y, KIM B Y, LEE K. 3D face model reconstruction from single 2D frontal image[C] //Proc of International Conference on Virtual Reality Continuum and Its Applications in Industry. 2009:95-99.
[5] KEMELMACHER-SHLIZERMAN I, BASRI R. 3D face reconstruction from a single image using a single reference face shape[J] . IEEE Trans on Pattern Analysis and Machine Intelligence, 2011, 33(2):394-405. [6] WANG Sheng-ming. Research of 3D face reconstruction based on one image[C] //Proc of International Conference on Image Analysis and Signal Processing. 2011:533-537.
[7] SONG Ming-li, TAO Da-cheng, HUANG Xiao-qin, et al. Three-dimensional face reconstruction from a single image by a coupled RBF network[J] . IEEE Trans on Image Processing, 2012, 21(5):2887-2897.
[8] 龚勋, 王国胤. 基于特征点的三维人脸形变模型[J] . 软件学报, 2009, 20(3):724-733. 〓
[9] YANG Meng, ZHANG Lei, SHIU S C K, et al. Robust kernel representation with statistical local features for face recognition[J] . IEEE Trans on Neural Networks and Learning Systems, 2013, 24(6):901-910.
[10] LUXIMON Y, BALL R, JUSTICE L. The 3D Chinese head and face modeling[J] . Computer-Aided Design, 2012, 44(1):40-47.
[11] 肖业清, 侯进, 王献. 基于正面照的自适应三维头部网格模型研究[J] . 计算机仿真, 2013, 30(7):412-415.
收稿日期 2014/3/3
修回日期 2014/4/24
页码 1240-1243,1247
中图分类号 TP391.41
文献标志码 A