Research map / Point clouds / Point cloud classification
Box-Aware Feature Enhancement for Single Object Tracking on Point Clouds
Chaoda Zheng, Xu Yan, Jiantao Gao, Weibing Zhao, Wei Zhang, Zhen Li, Shuguang Cui
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 template BoxCloud to feature aggregation input (ablation) | BAT · KITTI Car (ablation) | Success: 58.9 -> 60.5 | Improved |
| add predicted search-area BoxCloud to aggregation input (ablation) | BAT · KITTI Car (ablation) | Success: 60.5 -> 56.7 | Hurt |
| add pairwise distances to aggregation input (ablation) | BAT · KITTI Car (ablation) | Success: 60.5 -> 58.4 | Hurt |
| add search-point coordinates to aggregation input (ablation) | BAT · KITTI Car (ablation) | Success: 60.5 -> 56.7 | Hurt |
| compare with BoxCloud instead of PointNet++ features in kNN (ablation) | BAT · KITTI Car (ablation) | Success: 57.1 -> 60.5 | Improved |
| apply BoxCloud comparison to P2B aggregation (ablation) | P2B · KITTI Car (ablation) | Success: 56.2 -> 58.3 | Improved |
| add BoxCloud features to P2B (BAT-Vanilla) (ablation) | P2B · KITTI Car (ablation) | Success: 56.2 -> 59.0 | Improved |
| add BAFF module on top of BAT-Vanilla (ablation) | BAT-Vanilla · KITTI Car (ablation) | Success: 59.0 -> 60.5 | Improved |
| use Euclidean-distance BoxCloud instead of offset BoxCloud (ablation) | BAT · KITTI Car (ablation) | Success: 58.8 -> 60.5 | Improved |
| use first GT plus previous result for template instead of first GT only (ablation) | BAT · KITTI Car (ablation) | Success: 51.8 -> 60.5 | Improved |
| use first GT plus previous result for template instead of all previous results (ablation) | BAT · KITTI Car (ablation) | Success: 55.8 -> 60.5 | Improved |
| use larger k in kNN grouping (up to all 64 template points) (ablation) | BAT · KITTI Car (ablation) | Success | No clear effect |
Closest papers by meaning
- TrackAny3D: Transferring Pretrained 3D Models for Category-unified 3D Point Cloud Tracking (ICCV 2025 · similarity 0.89)
- Exploring Point-BEV Fusion for 3D Point Cloud Object Tracking With Transformer (TPAMI 2024 · similarity 0.89)
- An Effective Motion-Centric Paradigm for 3D Single Object Tracking in Point Clouds (TPAMI 2023 · similarity 0.89)
- Modeling Continuous Motion for 3D Point Cloud Object Tracking (AAAI 2024 · similarity 0.89)
- DiffuBox: Refining 3D Object Detection with Point Diffusion (NeurIPS 2024 · similarity 0.86)
- Towards Category Unification of 3D Single Object Tracking on Point Clouds (ICLR 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.
