| 基于三维模型的人脸姿态估计 |
| |
| 引用本文： | 牛晓霞.基于三维模型的人脸姿态估计[J].微处理机,2014(6):62-65. |
| |
| 作者姓名： | 牛晓霞 |
| |
| 作者单位： | 燕山大学信息科学与工程学院,秦皇岛066004 |
| |
| 基金项目： | 秦皇岛市科学技术研究与发展指导计划资助项目（2012021A048） |
| |
| 摘 要： | 提出了一种基于三维模型的人脸姿态估计方法。首先根据人脸特征点重建出稀疏的三维人脸模型,然后基于三维模型采用线性回归的方法对人脸姿态进行初步估计,确定姿态范围,再对估计结果进行修正,从而对人脸姿态进行精确估计。实验表明,该方法具有较好的估计效果,提高了姿态估计精度。
|
| 关 键 词： | 三维模型 线性回归 人脸姿态估计 稀疏模型 |
Face Pose Estimation Based on 3D Model |
| |
| Affiliation: | NIU Xiao - xia (College of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China) |
| |
| Abstract: | The face pose estimation method based on 3D model is proposed. First, the spare face model is built according to the feature points of the face. Then, based on the 3D face model, linear regression is used to estimate the face pose preliminarily and the 3D face pose is revised to acquire the accurate pose value. Finally, the face pose is estimated accurately. The experiment results show that the estimation method can obtain satisfying result and achieve better accuracy. |
| |
| Keywords: | 3D models Linear regression Face pose estimation Sparse model |
| 本文献已被 维普 等数据库收录！ |
|
