Research map / 3D vision / Face and shape
PlaneFormers: From Sparse View Planes to 3D Reconstruction
Samir Agarwala, Linyi Jin, Chris Rockwell, David F. Fouhey
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 appearance features to the plane token (ablation) | PlaneFormer · Matterport3D two-view | IPAA-90 (%): 26.9 -> 40.6 | Improved |
| add plane parameters to the plane token (ablation) | PlaneFormer · Matterport3D two-view | rotation mean error (deg): 25.92 -> 22.20 | Improved |
| add plane parameters to the plane token (ablation) | PlaneFormer · Matterport3D two-view | IPAA-90 (%): 35.2 -> 40.6 | Improved |
| add mask features to the plane token (ablation) | PlaneFormer · Matterport3D two-view | IPAA-90 (%): 34.5 -> 40.6 | Improved |
| add mask features to the plane token (ablation) | PlaneFormer · Matterport3D two-view | rotation mean error (deg): 21.21 -> 22.20 | Hurt |
| add transformer for inter-plane interaction (ablation) | PlaneFormer · Matterport3D two-view | IPAA-90 (%): 32.7 -> 40.6 | Improved |
| add transformer for inter-plane interaction (ablation) | PlaneFormer · Matterport3D two-view | translation mean error (m): 1.48 -> 1.19 | Improved |
| add camera residual (ablation) | PlaneFormer · Matterport3D two-view | translation mean error (m): 1.34 -> 1.19 | Improved |
| add camera residual (ablation) | PlaneFormer · Matterport3D two-view | IPAA-90 (%): 40.6 -> 40.6 | No clear effect |
| use PlaneFormer instead of Sparse Planes without continuous optimization | PlaneFormer · Matterport3D two-view | translation error <=1m (%): 56.5 -> 66.8 | Improved |
| use PlaneFormer instead of Sparse Planes | PlaneFormer · Matterport3D two-view | IPAA-90 (%): 28.1 -> 40.6 | Improved |
| use PlaneFormer instead of Sparse Planes (full) on rotation median | PlaneFormer · Matterport3D two-view | rotation median error (deg): 7.33 -> 5.96 | Improved |
Closest papers by meaning
- Towards In-the-wild 3D Plane Reconstruction from a Single Image (CVPR 2025 · similarity 0.89)
- AlphaTablets: A Generic Plane Representation for 3D Planar Reconstruction from Monocular Videos (NeurIPS 2024 · similarity 0.88)
- AirPlanes: Accurate Plane Estimation via 3D-Consistent Embeddings (CVPR 2024 · similarity 0.87)
- PLANA3R: Zero-shot Metric Planar 3D Reconstruction via Feed-forward Planar Splatting (NeurIPS 2025 · similarity 0.86)
- PlanarSplatting: Accurate Planar Surface Reconstruction in 3 Minutes (CVPR 2025 · similarity 0.86)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (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.
