Computer Engineering and Applications ›› 2021, Vol. 57 ›› Issue (15): 230-236.DOI: 10.3778/j.issn.1002-8331.2005-0073

Previous Articles     Next Articles

Visual Odometry with Self-Supervised Monocular Depth Estimation

CHENG Liming, CHEN Jianxin, CHEN Rui, CHEN Zhimin   

  1. College of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China
  • Online:2021-08-01 Published:2021-07-26

融合自监督单目图像深度估计的视觉里程计

成立明,陈建新,陈瑞,陈志敏   

  1. 南京邮电大学 通信与信息工程学院,南京 210003

Abstract:

Aiming at the obvious scale uncertainty problem of the direct method DSO (Direct Sparse Odometry), the impact of scale uncertainty on the system positioning accuracy is analyzed, and using a deep learning network that performs depth prediction on a single image is proposed. For the time-consuming problem of the back-end, a Preconditioned Conjugate Gradient(PCG) algorithm is utilized to accelerate the back-end solution. Compared with ORB-SLAM2, DSO, LDSO on the KITTI public dataset, the accuracy of localization of the system is significantly improved.

Key words: depth estimation, visual odometry, scale, preconditioned conjugate gradient, localization

摘要:

针对直接法DSO(Direct Sparse Odometry)存在的明显的尺度不确定性问题,对尺度不确定性给系统定位精度带来的影响进行分析,提出将对单幅图像进行深度估计的深度学习网络和DSO相结合的融合算法;针对DSO后端耗时问题,提出运用预处理共轭梯度(Preconditioned Conjugate gradient,PCG)算法优化后端求解部分。在KITTI公开数据集上与ORB-SLAM2、DSO、LDSO进行对比测试,系统的定位精度得到显著提高。

关键词: 深度估计, 视觉里程计, 尺度, 预处理共轭梯度, 定位