Research map / 3D vision / Depth and stereo
Spatially-Varying Outdoor Lighting Estimation From Intrinsics
Yongjie Zhu, Yinda Zhang, Si Li, Boxin Shi
Check 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 |
|---|---|---|---|
| combine intrinsic cues and diffuse-convolution guidance for global sky estimation | SOLID-Net · SOLID-Img test | sun azimuth error: 22.3^{\circ} -> 12.6^{\circ} | Improved |
| combine intrinsic cues and diffuse-convolution guidance for global sky estimation | SOLID-Net · SOLID-Img test | sun elevation error: ${11.0}^{\circ} -> 8.5^{\circ} | Improved |
| use SOLID-Net instead of LENet_sky | global lighting estimation · Real outdoor test | sun azimuth error: 37.5^\circ -> 28.9^\circ | Improved |
| use SOLID-Net instead of NeurIllum | real outdoor local lighting estimation · Real outdoor local lighting | SSIM: 0.228 -> 0.235 | Improved |
| use SOLID-Net instead of NeurIllum | real outdoor local lighting estimation · Real outdoor local lighting | MAE: 0.233 -> 0.203 | Improved |
| add a warped shadow panorama to the warped color panorama (ablation) | P-Net · SOLID-Img test local lighting | SSIM: 0.689 -> 0.736 | Improved |
| add the estimated sky map to the warped color panorama (ablation) | P-Net · SOLID-Img test local lighting | SSIM: 0.689 -> 0.793 | Improved |
| use the estimated sky map instead of the relighted Lambertian input (ablation) | P-Net with warped color input · SOLID-Img test local lighting | MAE: 0.618 -> 0.531 | Improved |
| add warped shadow panorama to warped color and sky inputs (ablation) | P-Net · SOLID-Img test local lighting | SSIM: 0.793 -> 0.803 | Improved |
| add warped shadow panorama to warped color and sky inputs (ablation) | P-Net · SOLID-Img test local lighting | MAE: 0.531 -> 0.523 | Improved |
| add the tonemapped SSIM loss (ablation) | P-Net · SOLID-Img test local lighting | SSIM: 0.760 -> 0.798 | Improved |
| add the tonemapped SSIM loss (ablation) | P-Net · SOLID-Img test local lighting | MAE: 0.564 -> 0.552 | Improved |
Closest papers by meaning
- Estimating Spatially-Varying Lighting in Urban Scenes with Disentangled Representation (ECCV 2022 · similarity 0.91)
- All-Weather Deep Outdoor Lighting Estimation (CVPR 2019 · similarity 0.90)
- Deep Sky Modeling for Single Image Outdoor Lighting Estimation (CVPR 2019 · similarity 0.90)
- Deep Outdoor Illumination Estimation (CVPR 2017 · similarity 0.89)
- Spatio-Temporal Outdoor Lighting Aggregation on Image Sequences Using Transformer Networks (IJCV 2022 · similarity 0.89)
- Complementary Intrinsics From Neural Radiance Fields and CNNs for Outdoor Scene Relighting (CVPR 2023 · similarity 0.89)
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.
