Research map / 3D vision / Depth and stereo
PhyIR: Physics-Based Inverse Rendering for Panoramic Indoor Images
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 |
|---|---|---|---|
| Use the more complete SVBRDF representation and physics-based renderer | PhyIR · FutureHouse test set | re-render MSE: 0.0108 -> 0.0061 | Helped |
| Add circular padding (ablation) | PhyIR material estimation · FutureHouse | base color MSE: 0.955 -> 0.940 | Helped |
| Add circular padding (ablation) | PhyIR material estimation · FutureHouse | roughness MSE: 2.037 -> 1.934 | Helped |
| Jointly train the full pipeline (ablation) | PhyIR material estimation · FutureHouse | base color MSE: 0.940 -> 0.926 | Helped |
| Add the trainable guided filter module GSNet (ablation) | PhyIR material estimation · FutureHouse | base color MSE: 0.926 -> 0.902 | Helped |
| Add the spatial-coherence loss (ablation) | PhyIR lighting estimation · FutureHouse | SSIM: 0.6150 -> 0.6169 | Helped |
| Add the spatial-coherence loss (ablation) | PhyIR lighting estimation · FutureHouse | re-render error (MSE): 0.0583 -> 0.0714 | Hurt |
| Add the physics-based re-render loss (ablation) | PhyIR lighting estimation · FutureHouse | re-render error (MSE): 0.0714 -> 0.0060 | Helped |
| Jointly train the full pipeline (ablation) | PhyIR lighting estimation · FutureHouse | SSIM: 0.6124 -> 0.6169 | Helped |
| Jointly train the full pipeline (ablation) | PhyIR lighting estimation · FutureHouse | re-render error (MSE): 0.0060 -> 0.0061 | Hurt |
| Use PhyIR instead of InvIndoor | panoramic lighting estimation · Real panoramic lighting dataset | mirror silver sphere relighting MAE: 0.2407 -> 0.1117 | Helped |
| Use PhyIR instead of UniFuse | depth estimation · 3D60 | RMSE: 0.1968 -> 0.2367 | Hurt |
Closest papers by meaning
- IRIS: Inverse Rendering of Indoor Scenes from Low Dynamic Range Images (CVPR 2025 · similarity 0.89)
- Inverse Rendering of Glossy Objects via the Neural Plenoptic Function and Radiance Fields (CVPR 2024 · similarity 0.88)
- VMINer: Versatile Multi-view Inverse Rendering with Near- and Far-field Light Sources (CVPR 2024 · similarity 0.88)
- PBR-NeRF: Inverse Rendering with Physics-Based Neural Fields (CVPR 2025 · similarity 0.87)
- RobIR: Robust Inverse Rendering for High-Illumination Scenes (NeurIPS 2024 · similarity 0.86)
- NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction (CVPR 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.