Research map / Point clouds / Point cloud classification
P2P: Part-to-Part Motion Cues Guide a Strong Tracking Framework for LiDAR Point Clouds
Jiahao Nie, Fei Xie, Sifan Zhou, Xueyi Zhou, Dong‐Kyu Chae, Zhiwei He
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 spatial motion modeling (ablation) | P2P-point · KITTI Mean | Success: 55.3 -> 66.2 | Improved |
| add part-to-part fusion (ablation) | P2P-point · KITTI Mean | Success: 59.4 -> 66.2 | Improved |
| add part-to-part fusion (vs random fusion of non-corresponding parts) (ablation) | P2P-voxel · KITTI Mean | Success: 48.76 -> 71.7 | Improved |
| share weights between feature extractors (ablation) | P2P-point · KITTI Mean | Success: 60.6 -> 66.2 | Improved |
| share weights between feature extractors (ablation) | P2P-voxel · KITTI Mean | Success: 69.0 -> 71.7 | Improved |
| feature extraction then fusion instead of fusion then extraction (ablation) | P2P-voxel · KITTI Mean | Success: 70.9 -> 71.7 | Improved |
| use proposed data augmentation instead of M2Track augmentation (ablation) | P2P-point · KITTI Mean | Success: 64.0 -> 66.2 | Improved |
| use proposed data augmentation instead of M2Track augmentation (ablation) | P2P-voxel · KITTI Mean | Success: 68.9 -> 71.7 | Improved |
| use PointNet instead of PointNet++ backbone (ablation) | P2P-point · KITTI Mean | Success: 38.4 -> 66.2 | Improved |
| use PointNet instead of DGCNN backbone (ablation) | P2P-point · KITTI Mean | Success: 63.5 -> 66.2 | Improved |
| use VoxelNet instead of PointPillars backbone (ablation) | P2P-voxel · KITTI Mean | Success: 68.1 -> 71.7 | Improved |
| add box size information (ablation) | P2P-point · KITTI Mean | Success: 66.2 -> 68.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CXTrack — CXTrack: Improving 3D Point Cloud Tracking With Contextual Information (CVPR 2023) · better on 1 of 1 shared result
- PTTR++ — PTTR: Relational 3D Point Cloud Object Tracking With Transformer (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- An Effective Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds (TPAMI 2023 · similarity 0.92)
- Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds (CVPR 2022 · similarity 0.92)
- Modeling Continuous Motion for 3D Point Cloud Object Tracking (AAAI 2024 · similarity 0.88)
- UAWTrack: Universal 3D Single Object Tracking in Adverse Weather (AAAI 2025 · similarity 0.87)
- M3SOT: Multi-Frame, Multi-Field, Multi-Space 3D Single Object Tracking (AAAI 2024 · similarity 0.87)
- ICP-Flow: LiDAR Scene Flow Estimation with ICP (CVPR 2024 · 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.
