Here To Stay: A Quantitative Comparison of Virtual Object Stability in Markerless Mobile AR
Here To Stay: A Quantitative Comparison of Virtual Object Stability in Markerless Mobile AR
复制标题
DOI：
10.1109/cphs56133.2022.9804545
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
T. Scargill;Gopika Premsankar;Jiasi Chen;M. Gorlatova
中科院分区：
文献类型：
--
作者：
T. Scargill;Gopika Premsankar;Jiasi Chen;M. Gorlatova
移动增强现实(AR)具有实现人类与网络物理系统之间身临其境的自然交互的潜力。尤其是无标记AR，通过不依赖基准标记或预定义图像，为用户提供了极大的便利性和灵活性。然而，在无标记的智能手机AR中，由于不准确的场景理解，导致设备姿势跟踪的错误，经常发生不需要的虚拟对象移动。我们研究了可能影响虚拟对象稳定性的因素，设计了测量虚拟对象稳定性的实验，并对六种不同的用户动作和五种不同的智能手机配置进行了系统的量化表征。我们的研究表明，除了最简单的设置外，虚拟物体在所有环境中都存在明显的空间不稳定现象(即使在性能最好的智能手机上，位置误差也超过10厘米)，并强调了进一步增强基于智能手机的无标记AR的姿势跟踪算法的必要性。
Mobile augmented reality (AR) has the potential to enable immersive, natural interactions between humans and cyber-physical systems. In particular markerless AR, by not relying on fiducial markers or predefined images, provides great convenience and flexibility for users. However, unwanted virtual object movement frequently occurs in markerless smartphone AR due to inaccurate scene understanding, and resulting errors in device pose tracking. We examine the factors which may affect virtual object stability, design experiments to measure it, and conduct systematic quantitative characterizations across six different user actions and five different smartphone configurations. Our study demonstrates noticeable instances of spatial instability in virtual objects in all but the simplest settings (with position errors of greater than 10cm even on the best-performing smartphones), and underscores the need for further enhancements to pose tracking algorithms for smartphone-based markerless AR.