Full Text Views
29
Pose estimation is a basic requirement for the autonomous behavior of robots. In this article we present a robust and fast visual odometry method to obtain camera poses by using RGB-D images. We first propose a motion estimation method based on sparse geometric constraint and derive the analytic Jacobian of the geometric cost function to improve the convergence performance, then we use our motion estimation method to replace the tracking thread in ORB-SLAM for improving its runtime performance. Experimental results show that our method is twice faster than ORB-SLAM while keeping the similar accuracy.
Ruibin GUO
National University of Defense Technology
Dongxiang ZHOU
National University of Defense Technology
Keju PENG
National University of Defense Technology
Yunhui LIU
The Chinese University of Hong Kong
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Ruibin GUO, Dongxiang ZHOU, Keju PENG, Yunhui LIU, "Fast Visual Odometry Based Sparse Geometric Constraint for RGB-D Camera" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 1, pp. 214-218, January 2019, doi: 10.1587/transinf.2018EDL8119.
Abstract: Pose estimation is a basic requirement for the autonomous behavior of robots. In this article we present a robust and fast visual odometry method to obtain camera poses by using RGB-D images. We first propose a motion estimation method based on sparse geometric constraint and derive the analytic Jacobian of the geometric cost function to improve the convergence performance, then we use our motion estimation method to replace the tracking thread in ORB-SLAM for improving its runtime performance. Experimental results show that our method is twice faster than ORB-SLAM while keeping the similar accuracy.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDL8119/_p
Copy
@ARTICLE{e102-d_1_214,
author={Ruibin GUO, Dongxiang ZHOU, Keju PENG, Yunhui LIU, },
journal={IEICE TRANSACTIONS on Information},
title={Fast Visual Odometry Based Sparse Geometric Constraint for RGB-D Camera},
year={2019},
volume={E102-D},
number={1},
pages={214-218},
abstract={Pose estimation is a basic requirement for the autonomous behavior of robots. In this article we present a robust and fast visual odometry method to obtain camera poses by using RGB-D images. We first propose a motion estimation method based on sparse geometric constraint and derive the analytic Jacobian of the geometric cost function to improve the convergence performance, then we use our motion estimation method to replace the tracking thread in ORB-SLAM for improving its runtime performance. Experimental results show that our method is twice faster than ORB-SLAM while keeping the similar accuracy.},
keywords={},
doi={10.1587/transinf.2018EDL8119},
ISSN={1745-1361},
month={January},}
Copy
TY - JOUR
TI - Fast Visual Odometry Based Sparse Geometric Constraint for RGB-D Camera
T2 - IEICE TRANSACTIONS on Information
SP - 214
EP - 218
AU - Ruibin GUO
AU - Dongxiang ZHOU
AU - Keju PENG
AU - Yunhui LIU
PY - 2019
DO - 10.1587/transinf.2018EDL8119
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 2019
AB - Pose estimation is a basic requirement for the autonomous behavior of robots. In this article we present a robust and fast visual odometry method to obtain camera poses by using RGB-D images. We first propose a motion estimation method based on sparse geometric constraint and derive the analytic Jacobian of the geometric cost function to improve the convergence performance, then we use our motion estimation method to replace the tracking thread in ORB-SLAM for improving its runtime performance. Experimental results show that our method is twice faster than ORB-SLAM while keeping the similar accuracy.
ER -