MagicVO: An End-to-End Hybrid CNN and Bi-LSTM Method for Monocular Visual Odometry
Authors:
Open Access
- 2 July 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 7, 94118-94127
Abstract
For the robotic positioning and navigation, visual odometry (VO) system is widely used. However, the errors of the traditional VO accumulate when the robot moves. Besides, this paper proposes a new framework to solve the problem of monocular visual odometry, called MagicVO. Based on Convolutional Neural Network (CNN) and Bi-directional LSTM (Bi-LSTM), MagicVO outputs a 6-DoF absolute-scale pose at each position of the camera with a sequence of continuous monocular images as input. It does not only utilize the outstanding performance of CNN in extracting the rich features of image frames fully but also learns the geometric relationship from image sequences pre and post through Bi-LSTM to get a more accurate prediction. A pipeline of the MagicVO is shown in Fig.1. The MagicVO is an end-to-end system, and the results of experiments on the KITTI dataset and the ETH dataset show that MagicVO has a better performance than traditional VO systems in the accuracy of pose and the generalization ability.Keywords
Funding Information
- Fundamental Research Funds for the Central Universities (2019PTB-010)
This publication has 25 references indexed in Scilit:
- Exploring Representation Learning With CNNs for Frame-to-Frame Ego-Motion EstimationIEEE Robotics and Automation Letters, 2015
- ORB-SLAM: A Versatile and Accurate Monocular SLAM SystemIEEE Transactions on Robotics, 2015
- Fast relocalisation and loop closing in keyframe-based SLAM2014
- Semi-dense Visual Odometry for a Monocular Camera2013
- Improving the Agility of Keyframe-Based SLAM2008
- Parallel Tracking and Mapping for Small AR Workspaces2007
- A Study of the Rao-Blackwellised Particle Filter for Efficient and Accurate Vision-Based SLAMInternational Journal of Computer Vision, 2007
- Real-Time and Robust Monocular SLAM Using Predictive Multi-resolution Descriptors2006
- Visual odometry2004
- Real-time simultaneous localisation and mapping with a single camera2003