.png){kind=icon}
.png){kind=icon}
Return
Frame-; -Frame Visual Odometry Estimation Network With Error Relaxation Method
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2022.3214823.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Estimating frame-;
-frame (F2F) visual odometry with monocular images has significant problems of propagated accumulated drift. We propose a learning-based approach for F2F monocular visual odometry estimation with novel and simple methods that consider the coherence of camera trajec;
ries without any post-processing. The proposed network consists of two stages: initial estimation and error relaxation. In the first stage, the network learns disparity images;
extract features and predicts relative camera pose between adjacent two frames through the attention, rotation, and translation networks. Then, loss functions are proposed in the error relaxation stage;
reduce the local drift, increasing consistency under dynamic driving scenes. Moreover, our skip-ordering scheme shows the effectiveness of dealing with sequential data. Experiments with the KITTI benchmark dataset show that our proposed network outperforms other approaches with higher and more stable performance.
Keywords:
Cameras
Robot vision systems
Feature extraction
Visual odometry
Estimation
Three-dimensional displays
Neural networks
Deep learning
Supervised learning
Trajec
ry tracking
Deep neural network
visual odometry
camera pose
odometry drift
camera trajec
ry
Journal
IF:
3.6
Papers:
9.8W
Citations:
29.4W