Research map / 3D vision / Depth and stereo
Learning Monocular Visual Odometry via Self-Supervised Long-Term Modeling
Yuliang Zou, Pan Ji, Quoc-Huy Tran, Jia-Bin Huang, Manmohan Chandraker
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add one-layer ConvLSTM to the pose network (ablation) | pose network baseline · KITTI Odometry Seq 09 | relative translation error (%): 7.55 -> 5.59 | Improved |
| add one-layer ConvLSTM to the pose network (ablation) | pose network baseline · KITTI Odometry Seq 10 | relative translation error (%): 10.43 -> 7.23 | Improved |
| add second ConvLSTM layer (ablation) | one-layer ConvLSTM · KITTI Odometry Seq 09 | relative translation error (%): 5.59 -> 4.23 | Improved |
| add second ConvLSTM layer (ablation) | one-layer ConvLSTM · KITTI Odometry Seq 10 | relative translation error (%): 7.23 -> 6.02 | Improved |
| add second-stage long-term training (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 09 | relative translation error (%): 4.23 -> 3.49 | Improved |
| add second-stage long-term training (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 10 | relative translation error (%): 6.02 -> 5.81 | Improved |
| add second-stage long-term training (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 09 | ATE RMSE (m): 9.77 -> 11.30 | Mixed |
| add cycle consistency constraint (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 09 | relative translation error (%): 5.02 -> 4.23 | Improved |
| add cycle consistency constraint (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 10 | relative translation error (%): 6.88 -> 6.02 | Improved |
| add pose features as input (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 09 | relative translation error (%): 5.64 -> 4.23 | Improved |
| add depth features as input (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 09 | relative translation error (%): 4.54 -> 4.23 | Improved |
| add memory buffer (ablation) | two-layer ConvLSTM · KITTI Odometry Seq 09 | relative translation error (%): 5.12 -> 4.23 | Improved |
Closest papers by meaning
- Sequential Adversarial Learning for Self-Supervised Deep Visual Odometry (ICCV 2019 · similarity 0.91)
- A Global Occlusion-Aware Approach to Self-Supervised Monocular Visual Odometry (AAAI 2021 · similarity 0.91)
- Self-Supervised Deep Visual Odometry With Online Adaptation (CVPR 2020 · similarity 0.90)
- LEAP-VO: Long-term Effective Any Point Tracking for Visual Odometry (CVPR 2024 · similarity 0.87)
- Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning (CVPR 2024 · similarity 0.87)
- Leveraging Consistent Spatio-Temporal Correspondence for Robust Visual Odometry (AAAI 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
