Research map / Point clouds / Point cloud classification
An Effective Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds
Chaoda Zheng, Xu Yan, Haiming Zhang, Baoyuan Wang, Shenghui Cheng, Shuguang Robert Cui, Zhen Li
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 basic motion augmentation (ablation) | BAT · KITTI Car | success (AUC %): 65.37 -> 63.59 | Hurt |
| add basic motion augmentation (ablation) | P2B · KITTI Car | success (AUC %): 56.20 -> 55.21 | Hurt |
| add basic motion augmentation (ablation) | M2-Track · KITTI Car | precision (AUC %): 77.12 -> 80.81 | Improved |
| add coin-flip test to motion augmentation (ablation) | M2-Track · KITTI Car | success (AUC %): 65.49 -> 69.93 | Improved |
| add temporal flipping alone to basic augmentation (ablation) | M2-Track · KITTI Car | success (AUC %): 65.49 -> 65.11 | No clear effect |
| add temporal flipping on top of coin-flip test (ablation) | M2-Track · KITTI Car | success (AUC %): 69.93 -> 71.14 | Improved |
| add box-aware feature enhancement (ablation) | M2-Track · KITTI Car | success (AUC %): 62.00 -> 65.49 | Improved |
| add box-aware feature enhancement (ablation) | M2-Track · NuScenes | success (AUC %): 53.68 -> 55.85 | Improved |
| add previous box refinement (ablation) | M2-Track · KITTI Car | success (AUC %): 64.23 -> 65.49 | Improved |
| add binary motion classification (ablation) | M2-Track · KITTI Car | success (AUC %): 65.74 -> 65.49 | Mixed |
| add stage II (ablation) | M2-Track · KITTI Car | success (AUC %): 61.29 -> 65.49 | Improved |
| use 3 frames in multi-frame ensembling instead of 2 (ablation) | M2-Track* · KITTI Car | success (AUC %): 71.14 -> 74.65 | Improved |
Closest papers by meaning
- Modeling Continuous Motion for 3D Point Cloud Object Tracking (AAAI 2024 · similarity 0.92)
- M3SOT: Multi-Frame, Multi-Field, Multi-Space 3D Single Object Tracking (AAAI 2024 · similarity 0.90)
- UAWTrack: Universal 3D Single Object Tracking in Adverse Weather (AAAI 2025 · similarity 0.89)
- Towards Category Unification of 3D Single Object Tracking on Point Clouds (ICLR 2024 · similarity 0.89)
- 3DSFLabelling: Boosting 3D Scene Flow Estimation by Pseudo Auto-labelling (CVPR 2024 · similarity 0.88)
- SeMoLi: What Moves Together Belongs Together (CVPR 2024 · similarity 0.88)
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.
