| 基于李群代数表达帧间位姿变化矩阵的3D视觉跟踪研究 |
| |
| 引用本文： | 张磊,张兴国.基于李群代数表达帧间位姿变化矩阵的3D视觉跟踪研究[J].江西师范大学学报(自然科学版),2012,0(5):466-471. |
| |
| 作者姓名： | 张磊 张兴国 |
| |
| 作者单位： | 南通大学机械工程学院, 江苏 南通 226019 |
| |
| 基金项目： | 江苏省高校自然科学基金(11KJB460007);南通市应用研究计划(K2010001)资助项目 |
| |
| 摘 要： | 基于 RAPiD 边缘跟踪方法的思想,采用以帧间位姿变化更新目标物绝对位姿值的办法,引入李群代数表达帧间位姿变化矩阵;然后将帧间位姿变化矩阵参数的求解问题转换为线性最小二乘的最优求解问题.在李群代数的表达方式下,最小二乘的最优问题可较容易获得线性化表达,方便了求解.实验显示了该算法的3D跟踪效果与目标物位姿值的估计结果.
|
| 关 键 词： | 视觉跟踪 3维 最小二乘问题 位姿估计 |
3D Visual Tracking Based on the SE(3) Group Representation of Interframe Pose Transform Matrix |
| |
| ZHANG Lei,ZHANG Xing-guo.3D Visual Tracking Based on the SE(3) Group Representation of Interframe Pose Transform Matrix[J].Journal of Jiangxi Normal University (Natural Sciences Edition),2012,0(5):466-471. |
| |
| Authors: | ZHANG Lei ZHANG Xing-guo |
| |
| Institution: | (School of the Mechanical Engineering,Nantong University,Nantong Jiangsu 226019,China) |
| |
| Abstract: | The SE(3) group is introduced for representing the interframe pose transform which can be used for updating the absolute pose of the object. Then, the 3D tracking problem can be equivalent to the least square optimization problem on parameter solving of the interframe pose transform. Owe to the SE(3) group representation, simpler linear representation can be used in optimization process, which pave the way for parameter solving. Experiments show the 3D tracking results of an object and the related pose estimation. |
| |
| Keywords: | visual tracking 3D least square problem pose estimation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《江西师范大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《江西师范大学学报(自然科学版)》下载免费的PDF全文 |
