Research map / 3D vision / Human pose estimation
RCP-LO: A Relative Coordinate Prediction Framework for Generalizable Deep LiDAR Odometry
Chen Liu, Wen Li, Yongshu Huang, Minghang Zhu, Yuyang Yang, Dunqiang Liu, Sheng Ao, Cheng Wang
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 |
|---|---|---|---|
| predict relative coordinates instead of direct pose regression (ablation) | RCP-LO · KITTI odometry 07-10 mean | t_rel: 0.89 -> 0.72 | Improved |
| add GW-SVD instead of MLP pose regression (ablation) | RCP-LO · KITTI odometry 07-10 mean | t_rel: 0.82 -> 0.72 | Improved |
| add GW-SVD instead of MLP pose regression (ablation) | RCP-LO · KITTI odometry 07-10 mean | r_rel: 0.39 -> 0.46 | Hurt |
| add geometric/feature weighting to SVD (GW-SVD vs Vanilla-SVD) (ablation) | RCP-LO · KITTI odometry 07-10 mean | t_rel: 1.07 -> 0.72 | Improved |
| add GW-SVD weighting instead of Mask-SVD (ablation) | RCP-LO · KITTI odometry 07-10 mean | t_rel: 1.22 -> 0.72 | Improved |
| add diffusion module instead of cost-volume for refinement (ablation) | RCP-LO · KITTI odometry 07-10 mean | t_rel: 0.86 -> 0.72 | Improved |
| add diffusion module instead of cost-volume for refinement (ablation) | RCP-LO · KITTI odometry 07-10 mean | r_rel: 0.53 -> 0.46 | Improved |
| add hierarchical refinement from level 3 to level 0 (ablation) | RCP-LO · KITTI odometry 07-10 mean | t_rel: 4.42 -> 0.72 | Improved |
| use relative coordinate prediction framework instead of cost-volume + MLP baseline (ablation) | RCP-LO · Ford (trained on KITTI) | t_rel: 7.02 -> 3.17 | Improved |
| use relative coordinate prediction framework instead of cost-volume + MLP baseline (ablation) | RCP-LO · Oxford (trained on KITTI) | t_rel: 30.24 -> 7.33 | Improved |
| use relative coordinate prediction framework instead of cost-volume + MLP baseline (ablation) | RCP-LO · KITTI-360 (trained on KITTI) | t_rel: 1.96 -> 2.13 | Hurt |
| use relative coordinate prediction framework instead of cost-volume + MLP baseline (ablation) | RCP-LO · KITTI-360 (trained on KITTI) | r_rel: 1.18 -> 1.23 | Hurt |
Closest papers by meaning
- Efficient 3D Deep LiDAR Odometry (TPAMI 2022 · similarity 0.90)
- DiffLoc: Diffusion Model for Outdoor LiDAR Localization (CVPR 2024 · similarity 0.88)
- DiffLO: Semantic-Aware LiDAR Odometry with Diffusion-Based Refinement (CVPR 2025 · similarity 0.87)
- DeepPointMap: Advancing LiDAR SLAM with Unified Neural Descriptors (AAAI 2024 · similarity 0.86)
- ICP-Flow: LiDAR Scene Flow Estimation with ICP (CVPR 2024 · similarity 0.86)
- Unlocking Generalization Power in LiDAR Point Cloud Registration (CVPR 2025 · similarity 0.85)
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.
