Research map / Detection and segmentation / CNN features
Learning Pairwise Inter-Plane Relations for Piecewise Planar Reconstruction
Yiming Qian, Yasutaka Furukawa
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 the contact energy term to the optimization (ablation) | PlaneRCNN + Ours · ScanNet planar reconstruction | CCE (cm): 14.82 -> 9.25 | Improved |
| add the contact energy term to the optimization (ablation) | PlaneAE + Ours · ScanNet planar reconstruction | CCE (cm): 13.24 -> 11.59 | Improved |
| add the contact energy term to the optimization (ablation) | PlaneRCNN + Ours · ScanNet planar reconstruction | Normal error (deg): 11.38 -> 11.11 | Improved |
| add the contact energy term to the optimization (ablation) | PlaneAE + Ours · ScanNet planar reconstruction | Normal error (deg): 9.38 -> 9.68 | Hurt |
| add inter-plane refinement to PlaneRCNN (ablation) | PlaneRCNN · ScanNet planar reconstruction | ROE (deg): 12.12 -> 10.06 | Improved |
| add inter-plane refinement to PlaneRCNN | PlaneRCNN · ScanNet planar reconstruction | Offset error (cm): 20.12 -> 20.09 | Improved |
| add inter-plane refinement to PlaneAE | PlaneAE · ScanNet planar reconstruction | Offset error (cm): 15.53 -> 15.85 | Hurt |
| add Segmentation-MPN refinement to PlaneRCNN | PlaneRCNN · Plane segmentation, contact line only | IoU%: 39.90 -> 48.96 | Improved |
| add Segmentation-MPN refinement to PlaneRCNN | PlaneRCNN · Plane segmentation, entire image | IoU%: 78.98 -> 80.90 | Improved |
| add refinement to PlaneAE | PlaneAE · Plane segmentation, entire image | IoU%: 69.86 -> 69.26 | Hurt |
| add refinement to PlaneAE | PlaneAE · Plane segmentation, contact line only | IoU%: 38.90 -> 41.72 | Improved |
| add RGB, depth and normal dot-product inputs to Orientation-CNN (mask only to all) (ablation) | Orientation-CNN · Inter-plane relationship, parallel | F1-score: 0.37 -> 0.60 | Improved |
Closest papers by meaning
- PlaneNet: Piece-Wise Planar Reconstruction From a Single RGB Image (CVPR 2018 · similarity 0.89)
- Single-Image Piece-Wise Planar 3D Reconstruction via Associative Embedding (CVPR 2019 · similarity 0.88)
- Learning Partonomic 3D Reconstruction from Image Collections (CVPR 2025 · similarity 0.84)
- AlphaTablets: A Generic Plane Representation for 3D Planar Reconstruction from Monocular Videos (NeurIPS 2024 · similarity 0.83)
- Towards In-the-wild 3D Plane Reconstruction from a Single Image (CVPR 2025 · similarity 0.83)
- AirPlanes: Accurate Plane Estimation via 3D-Consistent Embeddings (CVPR 2024 · similarity 0.83)
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.
