Monocular Visual Odometry using Learned Repeatability and Description
Monocular Visual Odometry using Learned Repeatability and Description
复制标题
DOI：
10.1109/icra40945.2020.9197406
复制
发表时间：
2020-05
期刊：
影响因子：
--
通讯作者：
Huaiyang Huang;Haoyang Ye;Yuxiang Sun;Ming Liu
中科院分区：
文献类型：
--
作者：
Huaiyang Huang;Haoyang Ye;Yuxiang Sun;Ming Liu
具有挑战性环境下单目视觉里程测量的鲁棒性和准确性受到广泛关注。在本文中，我们提出了一种利用学习重复性和描述的单目VO系统。在混合方案中，相机姿态最初以直接的方式在预测的可重复性地图上跟踪，然后使用逐块的3D-2D关联进行细化。局部特征参数化和自适应映射模块进一步增强了系统的不同功能。对具有挑战性的公共数据集进行了广泛的评估。在相机姿态估计上的竞争性能证明了该方法的有效性。对局部重建精度和运行时间的进一步研究表明，我们的系统能够保持一个健壮和轻量级的后端。
Robustness and accuracy for monocular visual odometry (VO) under challenging environments are widely concerned. In this paper, we present a monocular VO system leveraging learned repeatability and description. In a hybrid scheme, the camera pose is initially tracked on the predicted repeatability maps in a direct manner and then refined with the patch-wise 3D-2D association. The local feature parameterization and the adapted mapping module further boost different functionalities in the system. Extensive evaluations on challenging public datasets are performed. The competitive performance on camera pose estimation demonstrates the effectiveness of our method. Additional studies on the local reconstruction accuracy and running time exhibit that our system is capable of maintaining a robust and lightweight backend.