Research map / 3D vision / Depth and stereo
Deep Reflectance Maps
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 indirect normal-supervised approach instead of direct image-to-map prediction | Reflectance-map estimation · SMASHINg synthetic | MSE: .0019 -> .0018 | Helped |
| Use the indirect normal-supervised approach instead of direct image-to-map prediction | Reflectance-map estimation · SMASHINg synthetic | DSSIM: .0209 -> .0180 | Helped |
| Use the indirect normal-supervised approach instead of direct image-to-map prediction | Reflectance-map estimation · SMASHINg real | MSE: .0120 -> .0143 | Hurt |
| Use the indirect normal-supervised approach instead of direct image-to-map prediction | Reflectance-map estimation · SMASHINg real | DSSIM: .0976 -> .0991 | Hurt |
| Replace SparseNet with RBF interpolation (ablation) | Indirect reflectance-map estimation · SMASHINg synthetic | MSE: .0018 -> .0038 | Hurt |
| Replace SparseNet with RBF interpolation (ablation) | Indirect reflectance-map estimation · SMASHINg real | MSE: .0143 -> .0116 | Helped |
| Use ground-truth normals instead of estimated normals (ablation) | Indirect reflectance-map estimation · SMASHINg synthetic | MSE: .0018 -> .0008 | Helped |
| Use ground-truth normals instead of estimated normals (ablation) | Indirect reflectance-map estimation · SMASHINg synthetic | DSSIM: .0180 -> .0111 | Helped |
| Add a second normal loss on x,y coordinates (ablation) | Indirect normal-estimation network · SMASHINg synthetic normals | Mean: 14.3 -> 13.4 | Helped |
| Add a second normal loss on x,y coordinates (ablation) | Indirect normal-estimation network · SMASHINg synthetic normals | Median: 9.1 -> 8.2 | Helped |
| Add a second normal loss on x,y coordinates (ablation) | Indirect normal-estimation network · SMASHINg synthetic normals | RMSE: 20.6 -> 19.8 | Helped |
| Train the dual-loss network on upsampled normals (ablation) | Indirect normal-estimation network · SMASHINg synthetic normals | Mean: 13.4 -> 13.3 | Helped |
Closest papers by meaning
- SpecNeRF: Gaussian Directional Encoding for Specular Reflections (CVPR 2024 · similarity 0.88)
- Diffusion Reflectance Map: Single-Image Stochastic Inverse Rendering of Illumination and Reflectance (CVPR 2024 · similarity 0.88)
- Real-Time Neural BRDF with Spherically Distributed Primitives (CVPR 2024 · similarity 0.85)
- Practical Measurements of Translucent Materials with Inter-Pixel Translucency Prior (CVPR 2024 · similarity 0.85)
- Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects (CVPR 2025 · similarity 0.85)
- Neural Directional Encoding for Efficient and Accurate View-Dependent Appearance Modeling (CVPR 2024 · 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.