Research map / Point clouds / Point cloud classification
PCAN: 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrieval
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 |
|---|---|---|---|
| Add context-aware point attention to NetVLAD aggregation | PointNetVLAD · Oxford | average recall @1%: 81.01 -> 83.81 | Helped |
| Add context-aware point attention to NetVLAD aggregation | PointNetVLAD · U.S. | average recall @1%: 77.83 -> 79.05 | Helped |
| Add context-aware point attention to NetVLAD aggregation | PointNetVLAD · R.A. | average recall @1%: 69.75 -> 71.18 | Helped |
| Add context-aware point attention to NetVLAD aggregation | PointNetVLAD · B.D. | average recall @1%: 65.30 -> 66.82 | Helped |
| Add contextual information to the attention network (ablation) | Point Attention Network (PAN) · Oxford | average recall @1%: 81.90 -> 83.81 | Helped |
| Add contextual information to the attention network (ablation) | Point Attention Network (PAN) · U.S. | average recall @1%: 78.47 -> 79.05 | Helped |
| Add contextual information to the attention network (ablation) | Point Attention Network (PAN) · R.A. | average recall @1%: 69.69 -> 71.18 | Helped |
| Add contextual information to the attention network (ablation) | Point Attention Network (PAN) · B.D. | average recall @1%: 63.87 -> 66.82 | Helped |
| Use MSG with three scales rather than MRG (ablation) | PCAN · Oxford | average recall @1%: 81.35 -> 83.81 | Helped |
| Use MSG with three scales rather than MRG (ablation) | PCAN · U.S. | average recall @1%: 78.72 -> 79.05 | Hurt |
| Use MSG with three scales rather than MSG with scale 2 (ablation) | PCAN · U.S. | average recall @1%: 81.62 -> 79.05 | Hurt |
| Use MSG with three scales rather than MRG (ablation) | PCAN · R.A. | average recall @1%: 66.19 -> 71.18 | Helped |
Closest papers by meaning
- PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition (CVPR 2018 · similarity 0.89)
- Attentional ShapeContextNet for Point Cloud Recognition (CVPR 2018 · similarity 0.89)
- SOE-Net: A Self-Attention and Orientation Encoding Network for Point Cloud Based Place Recognition (CVPR 2021 · similarity 0.88)
- PointAttN: You Only Need Attention for Point Cloud Completion (AAAI 2024 · similarity 0.86)
- Riemann-based Multi-scale Attention Reasoning Network for Text-3D Retrieval (AAAI 2025 · similarity 0.84)
- Exploring Vision Semantic Prompt for Efficient Point Cloud Understanding (ICML 2025 · similarity 0.84)
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.