Research map / 3D vision / Face and shape
SALVe: Semantic Alignment Verification for Floorplan Reconstruction from Sparse Panoramas
John Lambert, Yuguang Li, Ivaylo Boyadzhiev, Lambert Wixson, Manjunath Narayana, Will Hutchcroft, James Hays, Frank Dellaert, Sing Bing Kang
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 |
|---|---|---|---|
| use SALVe-verified semantic alignment instead of keypoint SfM (vs OpenSfM) | SALVe · ZInD test set | median panoramas localized (%): 22.22 -> 57.10 | Improved |
| use SALVe-verified semantic alignment instead of keypoint SfM (vs OpenMVG) | SALVe · ZInD test set | median panoramas localized (%): 8.70 -> 57.10 | Improved |
| use SALVe instead of OpenMVG/OpenSfM for median translation error | SALVe (PGO + AA) · ZInD test set | median translation error (m): 0.10 -> 0.25 | Hurt |
| add vanishing-point axis alignment to spanning tree aggregation (ablation) | SALVe (Spanning Tree) · ZInD axis-alignment comparison (oracle layouts) | mean rotation error (deg): 5.41 -> 3.69 | Improved |
| add vanishing-point axis alignment to pose graph optimization (ablation) | SALVe (PGO) · ZInD axis-alignment comparison (oracle layouts) | median rotation error (deg): 1.53 -> 0.17 | Improved |
| use pose graph optimization instead of spanning tree without axis alignment (ablation) | SALVe · ZInD axis-alignment comparison (oracle layouts) | median rotation error (deg): 1.92 -> 1.53 | Improved |
| use pose graph optimization instead of spanning tree with axis alignment (ablation) | SALVe + AA · ZInD axis-alignment comparison (oracle layouts) | floorplan IoU: 0.56 -> 0.56 | No clear effect |
| add axis alignment to pose graph optimization (ablation) | SALVe (PGO) · ZInD axis-alignment comparison (oracle layouts) | floorplan IoU, mean: 0.56 -> 0.56 | No clear effect |
| replace layout-only rasterization input with RGB floor and ceiling textures in the verifier (ablation) | SALVe · ZInD test set | panoramas localized | Improved |
| add ceiling texture to floor texture input of verifier (ablation) | SALVe · ZInD test set | panoramas localized | Improved |
| use all of doors, windows and openings to create pose graph edges (ablation) | SALVe · ZInD test set | panoramas localized | Improved |
| use ResNet-152 instead of ResNet-50 as verifier backbone (ablation) | SALVe · ZInD test set | pairwise classification accuracy | Improved |
Closest papers by meaning
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction (NeurIPS 2026 · similarity 0.90)
- SLIBO-Net: Floorplan Reconstruction via Slicing Box Representation with Local Geometry Regularization (NeurIPS 2023 · similarity 0.89)
- Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries (CVPR 2023 · similarity 0.88)
- BADGR: Bundle Adjustment Diffusion Conditioned by Gradients for Wide-Baseline Floor Plan Reconstruction (CVPR 2025 · similarity 0.87)
- CAGE: Continuity-Aware edGE Network Unlocks Robust Floorplan Reconstruction (NeurIPS 2025 · similarity 0.86)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.86)
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.
