Research map / 3D vision / 3D generation
Intrinsic Image Fusion for Multi-View 3D Material Reconstruction
Peter Kocsis, Lukas Höllein, Matthias Nießner
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 parametric per-object affine texture model instead of per-texel mean aggregation (ablation) | IIF · Synthetic scenes | albedo PSNR: 13.43 -> 29.53 | Improved |
| add Laplace distribution matching on top of parametric model (ablation) | IIF · Synthetic scenes | albedo PSNR: 29.53 -> 30.79 | Improved |
| increase number of RGBX predictions from 1 to 16 (ablation) | IIF · Synthetic scenes | albedo PSNR: 29.62 -> 30.79 | Improved |
| increase predictions from 8 to 16 (ablation) | IIF · Synthetic scenes | albedo PSNR: 30.72 -> 30.79 | No clear effect |
| increase predictions from 1 to 16 (ablation) | IIF · Synthetic scenes | metallic L2 (x1000): 0.134 -> 0.134 | No clear effect |
| per-object mean aggregation instead of per-texel/parametric (ablation) | IRIS-style aggregation · Synthetic scenes | albedo SSIM: 0.753 -> 0.641 | Hurt |
| restrict path tracing to low-dimensional per-object parameters | IIF vs IRIS · Synthetic scenes | albedo PSNR: 15.86 -> 20.72 | Improved |
| IIF vs FIPT on metallic | IIF · Synthetic scenes | metallic L2: 0.006 -> 0.007 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- IRIS — IRIS: Inverse Rendering of Indoor Scenes from Low Dynamic Range Images (CVPR 2025) · better on 6 of 6 shared results
- NeILF++ — NeILF: Neural Incident Light Field for Physically-Based Material Estimation (ECCV 2022) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- IDArb: Intrinsic Decomposition for Arbitrary Number of Input Views and Illuminations (ICLR 2025 · similarity 0.88)
- Intrinsic Image Diffusion for Indoor Single-view Material Estimation (CVPR 2024 · similarity 0.88)
- LIRM: Large Inverse Rendering Model for Progressive Reconstruction of Shape, Materials and View-dependent Radiance Fields (CVPR 2025 · similarity 0.87)
- MaterialRefGS: Reflective Gaussian Splatting with Multi-view Consistent Material Inference (NeurIPS 2025 · similarity 0.87)
- MAGE : Single Image to Material-Aware 3D via the Multi-View G-Buffer Estimation Model (CVPR 2025 · similarity 0.87)
- Material Palette: Extraction of Materials from a Single Image (CVPR 2024 · similarity 0.87)
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.
