Research map / 3D vision / 3D generation
HorizonNet: Learning Room Layout With 1D Representation and Pano Stretch Data Augmentation
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 |
|---|---|---|---|
| replace dense O(HW) layout prediction with the 1D O(W) representation (ablation) | HorizonNet backbone on cuboid layout estimation · Cuboid layout test set | 3D IoU (%): 77.87 -> 80.65 | Helped |
| add Pano Stretch augmentation (ablation) | 1D layout model without RNN · Cuboid layout test set | 3D IoU (%): 80.65 -> 81.22 | Helped |
| add Pano Stretch augmentation (ablation) | 1D layout model with RNN · Cuboid layout test set | 3D IoU (%): 81.23 -> 83.74 | Helped |
| add RNN instead of fully connected layers (ablation) | 1D layout model without Pano Stretch · Cuboid layout test set | 3D IoU (%): 80.65 -> 81.23 | Helped |
| add RNN instead of fully connected layers (ablation) | 1D layout model with Pano Stretch · Cuboid layout test set | 3D IoU (%): 81.22 -> 83.74 | Helped |
| add Pano Stretch augmentation (ablation) | Dense O(HW) layout model · Cuboid layout test set | 3D IoU (%): 77.87 -> 79.64 | Helped |
| fine-tune on re-annotated non-cuboid rooms (ablation) | HorizonNet · Non-cuboid rooms; 13-fold validation on 65 re-annotated rooms | 3D IoU (%): 77.4 -> 82.5 | Helped |
| add Pano Stretch augmentation (ablation) | PSPNet for semantic segmentation · Stanford 2D3D; 20 training images | mIoU (%): 31.5 -> 33.2 | Helped |
| add Pano Stretch augmentation (ablation) | PSPNet for semantic segmentation · Stanford 2D3D; 50 training images | mIoU (%): 34.9 -> 36.1 | Helped |
| add Pano Stretch augmentation (ablation) | PSPNet for semantic segmentation · Stanford 2D3D; 100 training images | mIoU (%): 37.1 -> 38.4 | Helped |
| add Pano Stretch augmentation (ablation) | PSPNet for semantic segmentation · Stanford 2D3D; 200 training images | mIoU (%): 40.7 -> 41.9 | Helped |
| add Pano Stretch augmentation (ablation) | PSPNet for semantic segmentation · Stanford 2D3D; 500 training images | mIoU (%): 44.2 -> 44.3 | Helped |
Closest papers by meaning
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction (NeurIPS 2026 · similarity 0.89)
- PanoContext-Former: Panoramic Total Scene Understanding with a Transformer (CVPR 2024 · similarity 0.87)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.87)
- No More Ambiguity in 360deg Room Layout via Bi-Layout Estimation (CVPR 2024 · similarity 0.86)
- OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization (NeurIPS 2025 · similarity 0.85)
- BADGR: Bundle Adjustment Diffusion Conditioned by Gradients for Wide-Baseline Floor Plan Reconstruction (CVPR 2025 · 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.