Research map / 3D vision / Depth and stereo
Scale Recovery for Monocular Visual Odometry Using Depth Estimated With Deep Convolutional Neural Fields
Xiaochuan Yin, Xiangwei Wang, Xiaoguo Du, Qijun Chen
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 refinement with consecutive frames and egomotion to regressed depth (ablation) | CNF depth regression · KITTI depth (seq 00,08,09,10; within 20 m) | rel: 0.162 -> 0.112 | Improved |
| add refinement with consecutive frames and egomotion to regressed depth (ablation) | CNF depth regression · KITTI depth (seq 00,08,09,10; within 20 m) | log10: 0.062 -> 0.046 | Improved |
| add refinement with consecutive frames and egomotion to regressed depth (ablation) | CNF depth regression · KITTI depth (seq 00,08,09,10; within 20 m) | rms: 2.571 -> 2.047 | Improved |
| use learned-depth scale recovery instead of VISO2-M | monocular VO · KITTI odometry seq 10 | translation error (%): 25.20 -> 1.70 | Improved |
| use learned-depth scale recovery instead of VISO2-M | monocular VO · KITTI odometry seq 08 | translation error (%): 13.94 -> 2.22 | Improved |
| use learned-depth scale recovery instead of Song et al. | monocular VO · KITTI odometry seq 09 | translation error (%): 1.76 -> 4.14 | Hurt |
| use learned-depth scale recovery instead of Song et al. | monocular VO · KITTI odometry seq 08 | translation error (%): 2.37 -> 2.22 | Improved |
| use learned-depth scale recovery instead of Song et al. | monocular VO · KITTI odometry seq 10 | translation error (%): 2.12 -> 1.70 | Improved |
| use learned-depth scale recovery instead of P-CNN VO | monocular VO · KITTI odometry (avg) | translation error (%): 8.96 -> 2.68 | Improved |
| use learned-depth scale recovery instead of Song et al. | monocular VO · KITTI odometry (avg) | rotation error (deg/m): 0.0059 -> 0.0022 | Improved |
Closest papers by meaning
- Unsupervised Learning of Monocular Depth Estimation and Visual Odometry With Deep Feature Reconstruction (CVPR 2018 · similarity 0.90)
- Recurrent Neural Network for (Un-)Supervised Learning of Monocular Video Visual Odometry and Depth (CVPR 2019 · similarity 0.89)
- Deep Virtual Stereo Odometry: Leveraging Deep Depth Prediction for Monocular Direct Sparse Odometry (ECCV 2018 · similarity 0.89)
- D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry (CVPR 2020 · similarity 0.89)
- RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions (NeurIPS 2024 · similarity 0.86)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.86)
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.
