期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

基于主动外观模型的人脸合成技术

Active appearance model based expression synthesis method
下载PDF
导出
摘要 针对现有人脸表情合成中真实性欠佳或速度较慢的问题,提出了一种鲁棒的人脸表情合成方法。通过改进的主动外观模型(AAM)方法,对输入图像中人脸上的指定特征点进行自动检测跟踪,并以特征点为基础对人脸进行Delaunay三角分割、对齐和变形,在不同的人脸中建立有效的对应关系。通过图形计算器(GPU),对匹配的输入人脸和目标人脸进行加权渲染,得到新的人脸表情.为了提高人脸表情合成质量,采用色彩调整和边界过渡的操作,降低了不同人脸融合时带来的差异性.实验结果表明,与传统方法相比,该算法在计算速度上有较大的提高,在渲染效果上更加真实. A robust human expression synthesis method was proposed to resolve the problem that the current human expression methods are either too slow or non-photorealistic. Certain feature points on human faces from input images were automatically tracked by using an improved active appearance model （AAM）. Then the human face was Delaunay triangulated, aligned and warped by these points, and effective relationships between different human faces were described. Graphics processing unit （GPU） helped weighted rendering of the input face and the target one, thus resulting in a synthesized new facial expression. Automatic color adjustment and edge fusion were applied to enhance the rendering effect in order to reduce the dissimilarity between different faces. Experimental results showed that the method greatly improves the calculation speed and makes the results more photorealistic compared with the traditional methods.
作者 金城 沈学文 卜佳俊 陈纯 宋明黎
机构地区 复旦大学计算机科学与工程系 浙江传媒学院电子信息系 浙江大学计算机科学与技术学院
出处 《浙江大学学报（工学版）》 EI CAS CSCD 北大核心 2008年第7期1140-1144,共5页 Journal of Zhejiang University：Engineering Science
关键词 人脸表情合成 主动外观模型 Delaunay三角分割 图形计算器 加权渲染 human expression synthesis active appearance model （AAM） Delaunay triangulation graphics processing unit（GPU） weighted rendering
分类号 TP391.41 [自动化与计算机技术—计算机应用技术]
  • 相关文献

参考文献7

  • 1PARKE F. Computer generated animation of faces[C]// Proceedings of the ACM Annual Conference. Boston, USA: ACM, 1972: 451-457.
  • 2BEIER T, NEELY S. Feature-based image metamorphosis [J]. Computer Graphics. 1992, 26(2):35-42.
  • 3BREGLER C, COVELL M, SLANEY M. Video re write: driving visual speech with audio[C] // Proceedings of the ACM SIGGRAPH Conference on Computer Graphics. Los Angeles, USA: ACM, 1997:353-360.
  • 4LIU Z, SHAN Y, ZHANG Z. Expressive expression mapping with ratio images [C]// Proceedings of ACM SIGGRAPH 2001. New York, USA: ACM, 2001: 271 - 276.
  • 5SHASHUA A, RAVIV T R. The quotient image: class based re-rendering and recognition with Varying illuminations [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2001, 23(2): 129 - 139.
  • 6ZHANG Q, LIU Z, GUO B, et al. Geometry-driven photorealistic facial expression synthesis [C]// Proceedings of the 2003 ACM SIGGRAPH/Eurographics Symposium on Computer Animation. San Diego, USA: ACM, 2003:177 - 186.
  • 7金城,卜佳俊,陈华,陈纯,宋明黎,吴祺.自底向上的人脸特征点定位[J].浙江大学学报（工学版）,2008,42(5):794-799. 被引量：4

二级参考文献19

  • 1EDWARDS G, COOTES T, TAYLOR C. Face recognition using active appearance models [C]//Proceedings of the European Conference on Computer Vision (ECCV). Freiburg: Springer, 1998:581-595.
  • 2COOTES T, WALKER K, TAYLOR C. active appearance models [C]//Proceedings International Conference on Automatic Face Recognition (FG). Grenoble: IEEE, 2000: View-based of the IEEE and Gesture 227 - 232.
  • 3HOU X, LI S, ZHANG H, et al. Direct appearance models [C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Hawaii: IEEE, 2001: 828- 833.
  • 4LI S, YAN S, ZHANG H, et al. Multi-view face alignment using direct appearance models[C] //Proceedings of the IEEE International Conference on Automatic Face and Gesture Recognition ( FG ). Washington: IEEE, 2002:324 - 329.
  • 5LOY G, FLETCHER L, APOSTOLOFF N, et al. An adaptive fusion architecture for target tracking [C]// Proceedings of the IEEE International Conference on Automatic Face and Gesture Recognition (FG). Washington: IEEE, 2002:261-266.
  • 6YUILLE A, HALLINAN P, COHEN D. Feature extraction from faces using deformable templates [J]. International Journal of Computer Vision (IJCV), 1992, 8 (2) : 99 - 111.
  • 7PENTLAND A, MOGHADDAM B, STARNER T. View-based and modular eigenspaces for face recognition [C]//Proceedlngs of IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 1994: 84- 91.
  • 8COOTES T, TAYLOR C. Constrained active appearance models [C] //Proceedings of International Conference on Computer Vision (ICCV). Vancouver: IEEE, 2001:748 - 754.
  • 9ZHU Z, JI Q, FUJIMURA K. Combining Kalman filtering and mean shift for real time eye tracking under active IR illumination[C]//Proceedings of the International Conference on Pattern Recognition ( ICPR ). Quebec: IEEE, 2002: 318- 321.
  • 10PETAJAN E, GRAF H P. Robust face feature analysis for automatic speech reading and character animation [C]//Proceedings of the IEEE International Conference on Automatic Face and Gesture Recognition (FG). Killington: IEEE, 1996:357-362.

共引文献3

浙江大学学报（工学版）
2008年 第7期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部