Trajectory representation and landmark projection for continuous-time structure from motion
Trajectory representation and landmark projection for continuous-time structure from motion
复制标题
运动连续时间结构的轨迹表示和地标投影
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Per
中科院分区：
文献类型：
--
作者：
H. Ovrén;Per
本文重新审视了运动的连续时间结构问题，并介绍了一些改进收敛性和效率的扩展。用C2-连续样条的轨迹自然地将惯性测量作为所寻求的轨迹的衍生物。我们分析了SO（3）和R 3上的分裂样条插值以及SE（3）上的联合样条插值的行为，并表明后者隐含地耦合了平移和旋转的方向。这样的假设对于安装在机器人手臂上的摄像机来说是有意义的，但对于手持或身体安装的摄像机来说就不是了。我们在样条融合框架中的实验表明，在所有测试情况下，R 3和SO（3）上的分裂样条优于SE（3）样条。最后，我们研究了滚动快门相机上的地标重投影问题，并表明测试的重投影方法给出了类似的质量，而它们的计算负载变化了两个因素。
This paper revisits the problem of continuous-time structure from motion, and introduces a number of extensions that improve convergence and efficiency. The formulation with a C 2 -continuous spline for the trajectory naturally incorporates inertial measurements, as derivatives of the sought trajectory. We analyze the behavior of split spline interpolation on SO ( 3 ) and on R 3 , and a joint spline on SE ( 3 ) , and show that the latter implicitly couples the direction of translation and rotation. Such an assumption can make good sense for a camera mounted on a robot arm, but not for hand-held or body-mounted cameras. Our experiments in the Spline Fusion framework show that a split spline on R 3 and SO ( 3 ) is preferable over an SE ( 3 ) spline in all tested cases. Finally, we investigate the problem of landmark reprojection on rolling shutter cameras, and show that the tested reprojection methods give similar quality, whereas their computational load varies by a factor of two.