Research map / Point clouds / Point cloud classification
Beyond 3D Siamese Tracking: A Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use motion-centric tracking instead of appearance-matching Siamese tracking | 3D LiDAR single-object tracking · NuScenes Mean | Precision: +17.02 | Helped |
| use motion-centric tracking instead of appearance-matching Siamese tracking | 3D LiDAR single-object tracking · Waymo Open Dataset Mean | Precision: +21.73 | Helped |
| use motion-centric tracking instead of appearance-matching Siamese tracking | 3D LiDAR single-object tracking · KITTI Mean | Precision: ↑8.2 | Helped |
| add box-aware feature enhancement (ablation) | M²-Track · KITTI Car | Precision: ↓ 4.66 | Helped |
| add previous box refinement (ablation) | M²-Track · NuScenes Car | Precision: ↓ 3.15 | Helped |
| add binary motion classification (ablation) | M²-Track · KITTI Car | Success: ↑ 0.25 | Hurt |
| add binary motion classification (ablation) | M²-Track · NuScenes Car | Precision: ↓ 0.69 | Helped |
| add Stage-II motion-assisted shape completion (ablation) | M²-Track · KITTI Car | Success: ↓ 4.20 | Helped |
| add motion augmentation during training (ablation) | M²-Track · KITTI Car | Precision: ↑ 3.69 | Helped |
| add motion augmentation during training | BAT · KITTI Car | Precision: ↓ 1.89 | Hurt |
| use BAT as a re-tracker around M²-Track predictions | M²-Track · KITTI Car | Success: ↑ 3.73 | Helped |
| use P2B as a re-tracker around M²-Track predictions | M²-Track · KITTI Car | Precision: ↑ 0.99 | Helped |
Closest papers by meaning
- Modeling Continuous Motion for 3D Point Cloud Object Tracking (AAAI 2024 · similarity 0.91)
- 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 0.88)
- Instantaneous Perception of Moving Objects in 3D (CVPR 2024 · similarity 0.86)
- Robust 3D Tracking with Quality-Aware Shape Completion (AAAI 2024 · 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.