Research map / Detection and segmentation / CNN features
Re:PolyWorld - A Graph Neural Network for Polygonal Scene Parsing
Stefano Zorzi, Friedrich Fraundorfer
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 edge-aware attention with visual edge features to PolyWorld | PolyWorld · CrowdAI test | AP: 63.3 -> 67.2 | Improved |
| add edge-aware attention with visual edge features to PolyWorld | PolyWorld · CrowdAI test | C-IoU: 88.2 -> 89.7 | Improved |
| multi-instance vertex representation (K=4) for shared vertices | Re:PolyWorld · Structured3D floorplan | Room precision: 0.97 -> 0.99 | Improved |
| apply Re:PolyWorld to wireframe parsing instead of proposal-based HAWPv2 | Re:PolyWorld · Wireframe parsing | sAP10: 69.7 -> 50.2 | Hurt |
| apply Re:PolyWorld to wireframe parsing instead of proposal-based HAWPv2 | Re:PolyWorld · Wireframe parsing | R10: 63.1 -> 64.6 | Improved |
| use positional refinement offset (ablation) | Re:PolyWorld · Structured3D floorplan | polygon realism | Improved |
Closest papers by meaning
- PolyWorld: Polygonal Building Extraction With Graph Neural Networks in Satellite Images (CVPR 2022 · similarity 0.93)
- Joint Semantic-geometric Learning for Polygonal Building Segmentation (AAAI 2021 · similarity 0.88)
- Polygonal Building Extraction by Frame Field Learning (CVPR 2021 · similarity 0.86)
- PolyTransform: Deep Polygon Transformer for Instance Segmentation (CVPR 2020 · similarity 0.86)
- PolyhedronNet: Representation Learning for Polyhedra with Surface-attributed Graph (ICLR 2025 · similarity 0.84)
- PolarNeXt: Rethink Instance Segmentation with Polar Representation (CVPR 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.
