Research map / Detection and segmentation / CNN features
CompetitorFormer: Mitigating Query Conflicts for 3D Instance Segmentation via Competitive Strategy
Duanchu Wang, Junjie Yang, Haoran Gong, Jing Liu, Di Wang
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 Query Competition Layer (ablation) | Relation3D-style baseline · ScanNetV2 val | mAP: 62.2 -> 62.8 | Improved |
| add Relative Relationship Encoding (ablation) | baseline · ScanNetV2 val | mAP: 62.2 -> 62.7 | Improved |
| add Rank Cross-Attention (ablation) | baseline · ScanNetV2 val | mAP: 62.2 -> 62.6 | Improved |
| add QCL on top of RRE+RCA (ablation) | CompetitorFormer · ScanNetV2 val | mAP: 62.9 -> 63.4 | Improved |
| combine all three modules (ablation) | baseline · ScanNetV2 val | mAP: 62.2 -> 63.4 | Improved |
| increase RRE buckets L from 70 to 80 (ablation) | RRE · ScanNetV2 val | mAP: mAP 63.4 -> 63.1; mAP50 81.6 -> 81.8 | Mixed |
| use fewer RRE buckets with small width (L=50, v=0.010) (ablation) | RRE · ScanNetV2 val | mAP: 63.4 -> 62.6 | Hurt |
| add competition modeling | baseline · ScanNetV2 val | convergence: 62.0 mAP at epoch 300 vs baseline at epoch 450 | Improved |
| add competition modeling | Relation3D · ScanNet200 val | mAP: 31.6 -> 34.1 | Improved |
| add competition modeling | DCD · ScanNet++V2 val | mAP: 28.5 -> 34.1 | Improved |
Closest papers by meaning
- Relation3D : Enhancing Relation Modeling for Point Cloud Instance Segmentation (CVPR 2025 · similarity 0.86)
- Insightful Instance Features for 3D Instance Segmentation (CVPR 2025 · similarity 0.86)
- Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking (CVPR 2025 · similarity 0.86)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.85)
- Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior (CVPR 2024 · similarity 0.85)
- AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation (ICLR 2024 · 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.
