Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method
Estimating frame-to-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 trajectories without a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9919836/ |
| _version_ | 1811250128364765184 |
|---|---|
| author | Sangwon Hwang Myeongah Cho Yuseok Ban Kyungjae Lee |
| author_facet | Sangwon Hwang Myeongah Cho Yuseok Ban Kyungjae Lee |
| author_sort | Sangwon Hwang |
| collection | DOAJ |
| description | Estimating frame-to-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 trajectories 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 to 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 to 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. |
| first_indexed | 2024-04-12T15:58:37Z |
| format | Article |
| id | doaj.art-be8033102fa546efbf25e424825a5437 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-12T15:58:37Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-be8033102fa546efbf25e424825a54372022-12-22T03:26:16ZengIEEEIEEE Access2169-35362022-01-011010999411000210.1109/ACCESS.2022.32148239919836Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation MethodSangwon Hwang0https://orcid.org/0000-0001-5549-5846Myeongah Cho1https://orcid.org/0000-0001-9330-2785Yuseok Ban2https://orcid.org/0000-0003-1190-6863Kyungjae Lee3https://orcid.org/0000-0002-1529-3120School of Electrical and Electronic Engineering, Yonsei University, Seoul, South KoreaSchool of Electrical and Electronic Engineering, Yonsei University, Seoul, South KoreaSchool of Electronics Engineering, Chungbuk National University, Cheongju, South KoreaSchool of Artificial Intelligence, Yong In University, Yongin, South KoreaEstimating frame-to-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 trajectories 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 to 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 to 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.https://ieeexplore.ieee.org/document/9919836/Deep neural networkvisual odometrycamera poseodometry driftcamera trajectory |
| spellingShingle | Sangwon Hwang Myeongah Cho Yuseok Ban Kyungjae Lee Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method IEEE Access Deep neural network visual odometry camera pose odometry drift camera trajectory |
| title | Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method |
| title_full | Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method |
| title_fullStr | Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method |
| title_full_unstemmed | Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method |
| title_short | Frame-to-Frame Visual Odometry Estimation Network With Error Relaxation Method |
| title_sort | frame to frame visual odometry estimation network with error relaxation method |
| topic | Deep neural network visual odometry camera pose odometry drift camera trajectory |
| url | https://ieeexplore.ieee.org/document/9919836/ |
| work_keys_str_mv | AT sangwonhwang frametoframevisualodometryestimationnetworkwitherrorrelaxationmethod AT myeongahcho frametoframevisualodometryestimationnetworkwitherrorrelaxationmethod AT yuseokban frametoframevisualodometryestimationnetworkwitherrorrelaxationmethod AT kyungjaelee frametoframevisualodometryestimationnetworkwitherrorrelaxationmethod |