Research map / 3D vision / Gaussian splatting
MatSpray: Fusing 2D Material World Knowledge on 3D Geometry
Philipp Langsteiner, Jan-Niklas Dihlmann, Hendrik Lensch
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 Neural Merger (softmax fusion of per-view materials) on top of projected-average materials (ablation) | MatSpray · ablation subset relighting | PSNR: 25.555 -> 29.164 | Improved |
| add Neural Merger and 3D photometric optimization versus supervising only with 2D diffusion maps (ablation) | MatSpray · ablation subset relighting | PSNR: 24.809 -> 29.164 | Improved |
| optimize with 2D supervision only instead of simple projection averaging (ablation) | MatSpray · ablation subset relighting | PSNR: 25.555 -> 24.809 | Hurt |
| use MatSpray instead of IRGS for relighting | IRGS · ARIA synthetic (17 objects) | PSNR: 24.409 -> 27.282 | Improved |
| use MatSpray instead of extended R3DGS for base color | Ext. R3DGS · ARIA synthetic (17 objects) | PSNR: 18.360 -> 21.341 | Improved |
| use MatSpray for roughness instead of IRGS | IRGS · ARIA synthetic (17 objects) roughness | PSNR: 16.182 -> 15.331 (SSIM 0.744 -> 0.820) | Mixed |
| add metallic prediction via diffusion material lifting versus extended R3DGS | Ext. R3DGS · ARIA synthetic (17 objects) metallic | PSNR: 10.073 -> 27.202 | Improved |
| use MatSpray instead of IRGS on Stanford Orb relighting | IRGS · Stanford Orb | PSNR: 30.2 -> 31.2 (SSIM .97 -> .95) | Mixed |
| use MatSpray instead of IRGS on NeRF Synthetic | IRGS · NeRF Synthetic | PSNR: 23.5 -> 25.4 | Improved |
| choose DiffusionRenderer as 2D material predictor over Marigold and RGB-to-X (ablation) | MatSpray · material prediction | PSNR: about thirty percent higher PSNR | Improved |
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.
- IRGS — IRGS: Inter-Reflective Gaussian Splatting with 2D Gaussian Ray Tracing (CVPR 2025) · better on 8 of 9 shared results
- LightSwitch — LightSwitch: Multi-view Relighting with Material-guided Diffusion (ICCV 2025) · better on 3 of 3 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
- TexGaussian: Generating High-quality PBR Material via Octree-based 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- MaterialRefGS: Reflective Gaussian Splatting with Multi-view Consistent Material Inference (NeurIPS 2025 · similarity 0.90)
- Material Palette: Extraction of Materials from a Single Image (CVPR 2024 · similarity 0.89)
- OMG: Opacity Matters in Material Modeling with Gaussian Splatting (ICLR 2025 · similarity 0.89)
- Material Anything: Generating Materials for Any 3D Object via Diffusion (CVPR 2025 · similarity 0.88)
- MAGE : Single Image to Material-Aware 3D via the Multi-View G-Buffer Estimation Model (CVPR 2025 · similarity 0.88)
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.
