| 基于SIFT特征的视频内头部姿态估计 |
| |
| 引用本文： | 孙玉芬,刘钢.基于SIFT特征的视频内头部姿态估计[J].武汉理工大学学报,2011(1). |
| |
| 作者姓名： | 孙玉芬 刘钢 |
| |
| 作者单位： | 武汉理工大学计算机科学与技术学院; |
| |
| 基金项目： | 国家自然科学基金(60974094); 水路公路交通安全与装备教育部工程研究中心开放基金(WHUTERCTS2010A04); 中央高校基本科研业务费专项资金(2010-Ⅳ-085) |
| |
| 摘 要： | 基于对头部左右转动时脸部图像上点之间距离变化特性的分析,提出基于SIFT特征的单摄像头近距离视频图像序列内头部姿态估计算法。该方法首先在脸部图像上提取SIFT特征点,然后对两幅图像进行SIFT特征匹配,基于匹配的特征点之间距离的变化性质判断两幅图像内头部的相对转向与转动程度,得到头部姿态估计结果。在人脸库和视频图像上的实验表明,该方法使用简单,且具有较高的正确度。
|
| 关 键 词： | 头部姿态估计 SIFT特征 透视投影 |
SIFT Features Based Head Pose Estimation from Monocular Video Sequences |
| |
| SUN Yu-fen,LIU Gang.SIFT Features Based Head Pose Estimation from Monocular Video Sequences[J].Journal of Wuhan University of Technology,2011(1). |
| |
| Authors: | SUN Yu-fen LIU Gang |
| |
| Affiliation: | SUN Yu-fen,LIU Gang(School of Computer Science & Technology,Wuhan University of Technology,Wuhan 430063,China) |
| |
| Abstract: | The change of the distances between points on face images is distinctive when the face yaws.A head pose estimation algorithm using SIFT features is proposed based on this distinctive change pattern.The algorithm first computes SIFT features in face images.Then the features in two images are matched.The head pose in one image is estimated based on the change pattern of the distances between these feature points and the head pose in the other image.The experiments on face database and monocular video sequence... |
| |
| Keywords: | head pose estimation SIFT feature perspective projection |
| 本文献已被 CNKI 等数据库收录！ |
