| 基于多点模型的3D人脸姿态估计方法 |
| |
| 引用本文： | 胡步发,邱丽梅. 基于多点模型的3D人脸姿态估计方法[J]. 中国图象图形学报, 2008, 13(7): 1353-1358 |
| |
| 作者姓名： | 胡步发 邱丽梅 |
| |
| 作者单位： | 福州大学机械工程及自动化学院 |
| |
| 摘 要： | 改进传统的活动形状模型法,准确地提取人脸特征点后,针对人脸形状特性,使用人脸的多个特征点作为人脸模型,通过最小二乘法优化求解,精确估计3D人脸空间姿态。实验结果表明,新方法不仅可以获得稳定的姿态解,而且与同类方法比较具有良好的姿态估计精确度。
|
| 关 键 词： | 人脸姿态估计 活动形状模型 最优化 |
| 收稿时间： | 2006-12-30 |
| 修稿时间： | 2007-03-16 |
3D Face Pose Estimation Based on Multi points Model |
| |
| HU Bu-f,QIU Li-mei and HU Bu-f,QIU Li-mei. 3D Face Pose Estimation Based on Multi points Model[J]. Journal of Image and Graphics, 2008, 13(7): 1353-1358 |
| |
| Authors: | HU Bu-f QIU Li-mei HU Bu-f QIU Li-mei |
| |
| Affiliation: | (College of Mechanical Engineering and Automation, Fuzhou University, Fuzhou 350002, China) |
| |
| Abstract: | The improved method of active shape model is used to extract face feature points accurately.Then in view of the characteristic of face shape,taking some face feature points as the face model,3D space face pose is estimated via the least-squares method optimization.The experiment results show that the new method not only can achieve steady pose solution,but also has better estimation accuracy compared with the same kind methods. |
| |
| Keywords: | estimation of face pose active shape model(ASM) optimization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《中国图象图形学报》浏览原始摘要信息 |
|
点击此处可从《中国图象图形学报》下载免费的PDF全文 |
|
