Research map / 3D vision / NeRF and radiance fields
NEMTO: Neural Environment Matting for Novel View and Relighting Synthesis of Transparent Objects
Dongqing Wang, Tong Zhang, Sabine Süsstrunk
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 |
|---|---|---|---|
| replace analytical refraction with learned ray bending network (RBN) (ablation) | NEMTO · Synthetic novel view | PSNR: 21.758 -> 26.582 | Improved |
| replace analytical refraction with learned RBN (ablation) | NEMTO · Synthetic relighting | PSNR: 17.846 -> 25.147 | Improved |
| add refraction guiding loss L_rg (ablation) | NEMTO · Synthetic novel view | PSNR: 15.659 -> 26.582 | Improved |
| add refraction smoothness loss L_rs (ablation) | NEMTO · Synthetic novel view | PSNR: 21.623 -> 26.582 | Improved |
| add L_rs (ablation) | NEMTO · Synthetic relighting | PSNR: 19.026 -> 25.147 | Improved |
| use NEMTO with RBN instead of PhySG Disney BSDF appearance | inverse rendering · Synthetic relighting | PSNR: 15.412 -> 25.147 | Improved |
| use NEMTO instead of IDR for geometry | surface reconstruction · Synthetic meshes (Cow) | Chamfer L1 (1e-3): 11.66 -> 2.60 | Improved |
| use NEMTO instead of nerfacto on real captured transparent object | novel view synthesis · Real captured cat | PSNR: 24.67 -> 27.34 | Improved |
Closest papers by meaning
- Opti-NeuS: Neural Reconstruction for Dual-Layered Transparent and Opaque Objects (CVPR 2026 · similarity 0.89)
- Inverse Rendering of Glossy Objects via the Neural Plenoptic Function and Radiance Fields (CVPR 2024 · similarity 0.89)
- Differentiable Neural Surface Refinement for Modeling Transparent Objects (CVPR 2024 · similarity 0.88)
- From Transparent to Opaque: Rethinking Neural Implicit Surfaces with $\alpha$-NeuS (NeurIPS 2024 · similarity 0.88)
- A General Implicit Framework for Fast NeRF Composition and Rendering (AAAI 2024 · similarity 0.87)
- NeRFPrior: Learning Neural Radiance Field as a Prior for Indoor Scene Reconstruction (CVPR 2025 · 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.
